Traffic risk characterization and assessment method, device and equipment for urban road intersections

Abstract

The present invention proposes a method, device and equipment for characterizing and evaluating traffic risks at urban road intersections. A traffic risk index is constructed in combination with the frequency and severity of traffic conflicts; a smoke plume diffusion model is introduced to simulate the propagation and diffusion process of traffic risks at conflict points to the neighborhood, and the risk distribution of the entire intersection is characterized; based on the intersection risk distribution, a multi-dimensional hierarchical evaluation system for refined traffic risks at intersections is implemented, mainly including overall risks based on area, path traffic risks based on a single motion trajectory, and traffic mode risks based on multiple trajectories. The quantitative model for the overall traffic risk at intersections and the multi-dimensional hierarchical evaluation system constructed by the present invention enable risk assessment under single-path or group traffic modes through risk field simulation methods, which is expected to play an important supporting role in future urban risk analysis, and thus provide technical support for the development and construction of smart cars, smart transportation, safe travel and new smart cities.

Description

Technical Field

[0001] The present invention relates to the technical field of traffic risk assessment, and in particular to a method, device and equipment for characterizing and assessing traffic risks at urban road intersections. Background Art

[0002] Urban road intersections contain a large number of traffic conflicts and traffic risks due to their complex spatiotemporal dynamics and multimodal characteristics. In developed and developing countries, such as the United States, Norway, Japan, Singapore, and Thailand, road intersection traffic accidents account for approximately 20-45% of all reported traffic accidents (Golembiewsk et al., 2011; Ma et al., 2021; Tay et al., 2017; Elvik et al., 2005; Lord et al., 2007). Assessing the traffic risk of urban road intersections has always been a major means of identifying potential safety risks in intersection design (Essa et al., 2018), which is crucial to improving traffic conditions and ensuring the personal safety of road users (Nopadon Kronprasert et al., 2021).

[0003] In the past few decades, intersection traffic risk assessment has been a research focus in the field of traffic risk assessment. (Chen et al., 2020) tracked vehicles in the intersection area through computer vision technology, and used the TTC indicator to calculate the location and severity of potential conflict points in the area; (Wei et al., 2019) proposed a three-dimensional cube search algorithm to detect the proximity between trajectories to determine traffic conflicts and conflict severity; (Vittorio Astarita et al., 2019) proposed to evaluate the severity of the resulting potential collision based on accurate collision dynamics variables, thereby calculating the risk value of traffic conflicts; (Shi et al., 2018) proposed the concept of key risk indicators (KRI) and used mixed indicators to evaluate risk exposure. Based on the time-integrated collision time (TIT) and accident probability index (CPI) that can identify traffic risks before accidents, KPI expressions are developed to evaluate the severity and possibility of traffic risks at intersections; (Guo et al., 2016) evaluated the safety impact of unconventional external left-turn lanes at signalized intersections; (Justice et al., 2020) studied how the left-turn collision risk at signalized intersections changes with traffic conditions. The above methods only focus on traffic risks at conflict points. After a traffic conflict occurs in an actual intersection scenario, the passage of road users within a certain range of the conflict point will be affected, thereby increasing their traffic risks. In short, the impact of traffic conflict points will spread to the surrounding areas, and non-conflict point locations may also have a certain degree of risk, and existing methods cannot support risk calculations at any location. At the same time, from the perspective of risk quantification at intersections to guide road users' travel, the evaluation perspectives have three levels: intersection level, mode level, and path level. However, the above methods cannot take into account all aspects;

[0004] Therefore, realizing traffic risk assessment at any location and realizing multi-level traffic risk assessment at intersection level, mode level and path level are technical problems that need to be solved urgently. Summary of the invention

[0005] In order to solve the above technical problems, the present invention provides a method and corresponding device and electronic equipment for revealing the spatiotemporal dynamic risk distribution of intersections by establishing a multi-dimensional risk quantification model to provide quantitative reference for risk decision-making.

[0006] According to a first aspect of the present invention, the present invention provides a method for characterizing and evaluating traffic risks at urban road intersections, comprising the following steps:

[0007] The trajectory classification method of dynamic time normalization is used to classify the trajectories of moving objects in the intersection and extract the traffic mode;

[0008] The collision time is used as a key indicator to measure whether two vehicles have a traffic conflict, so as to identify the trajectory point where the traffic conflict occurs, namely the conflict point;

[0009] Detect the traffic mode to which the conflict point belongs, and classify the conflict point according to the traffic mode to which it belongs. Different traffic modes will generate different initial values ​​of risk diffusion to act on the risk diffusion simulation of a single object;

[0010] By integrating the frequency and severity of traffic conflicts, a traffic risk index R is constructed to quantify the traffic risk at the conflict point.

[0011] Combined with the Gaussian smoke plume diffusion model, a traffic risk diffusion model that conforms to the traffic movement mechanism is constructed;

[0012] Taking the conflict point as the diffusion source, the traffic risk diffusion model is used to perform traffic risk diffusion processing to obtain the traffic risk distribution of the intersection;

[0013] Calculate the diffusion area of ​​traffic risk inside the intersection to evaluate the traffic risk level at the intersection level;

[0014] The traffic risk change curve of the traffic trajectory in the intersection is calculated based on the traffic risk distribution of the intersection, and its integral is used to evaluate the traffic risk level at the path level;

[0015] Combined with the traffic risk level at the path level, the average of the traffic risk levels of all paths in the traffic mode is calculated to evaluate the traffic risk level at the traffic mode level.

[0016] Preferably, the step of using the trajectory classification method of dynamic time normalization to classify the trajectories of moving objects in the intersection and extracting the traffic mode includes:

[0017] Dynamic time normalization is used to calculate the similarity between the traffic trajectories inside the intersection, and the appropriate threshold is set to classify the trajectories. The classification result is sufficient to distinguish the traffic modes inside the intersection and extract the types of traffic modes in the intersection. Its function is to give different risk diffusion initial values ​​to different conflict points.

[0018] Preferably, the step of using the collision time as a key indicator to measure whether a traffic conflict occurs between two vehicles, so as to identify the trajectory point where the traffic conflict occurs, includes:

[0019] By calculating the collision time of all trajectory points in the trajectory and setting the collision time threshold, trajectory points with collision time higher than the threshold are considered conflict points and are screened out.

[0020] Preferably, the step of integrating the traffic conflict frequency and severity to construct a traffic risk index R to quantify the traffic risk of the conflict point includes:

[0021] With the help of the traffic conflict frequency Sum and severity SI in the traffic conflict technical concept, a risk index R is constructed;

[0022] The traffic risk of various conflict points is calculated based on the risk index R, and the formula is as follows:

[0023] R i =0.5×Sum i +0.5×SI i

[0024] In the formula, R i represents the traffic risk value of type i conflict point; Sum i represents the traffic conflict frequency of type i conflict point; SI i Indicates the severity of traffic conflict at type i conflict point.

[0025] Preferably, the step of combining the Gaussian smoke plume diffusion model to construct a traffic risk diffusion model that conforms to the traffic movement mechanism includes:

[0026] According to the Gaussian plume diffusion model, the risk diffusion ratio is used as the pollutant, and the process of pollutant diffusion to the surrounding area is used to simulate the traffic risk diffusion process;

[0027] The traffic risk index R is used as the diffusion index, the main diffusion direction is the movement speed direction, and the 180-degree angle interval in the speed direction is used as the diffusion interval. The traffic risk diffusion model is constructed as follows:

[0028]

[0029] In the formula, c(x,y) represents the traffic risk value at the position (x,y); δ represents the risk intensity of the diffusion source; μ represents the speed of the source object vehicle; σ y represents the diffusion coefficient.

[0030] Preferably, the step of calculating the diffusion area rate of the traffic risk inside the intersection includes:

[0031] Taking the risk value P as the threshold, the area greater than or equal to P is the traffic risk area, and the area less than P is the safe area. The traffic risk area is accumulated and the area is calculated to obtain the traffic risk diffusion area inside the intersection. The calculation formula is as follows:

[0032] S risk =∑s j if s j.risk ≥P

[0033] In the formula, S risk Represents the traffic risk diffusion area inside the intersection; s j represents the area of ​​region j; s j.riskrepresents the traffic risk value of area j; P represents the traffic risk threshold.

[0034] Preferably, the step of calculating the traffic risk change curve of the traffic trajectory in the intersection in combination with the traffic risk distribution of the intersection, and using the integral thereof to evaluate the traffic risk level of the path level comprises:

[0035] The traffic risk change curve of the traffic trajectory in the intersection is calculated according to the traffic risk distribution of the intersection. The traffic risk integral along the path is used to evaluate the traffic risk level of a single object when passing through the intersection. The calculation formula is as follows:

[0036]

[0037] Where RL l represents the traffic risk of trajectory l, represents the sum of the traffic risks of all trajectories in trajectory l, t represents the travel time of trajectory l, and α is the adjustment coefficient.

[0038] Preferably, the step of calculating the mean of the traffic risk levels of all paths in the traffic mode in combination with the traffic risk level at the path level to evaluate the traffic risk level at the traffic mode level comprises:

[0039] The traffic risk of a traffic mode or a certain turning mode is evaluated by using the mean of the traffic risk levels of the same type of traffic modes including trajectories. The calculation formula is as follows:

[0040]

[0041] In the formula, RM m represents the traffic risk of traffic mode m, represents the sum of traffic risks of all trajectories in mode m.

[0042] According to a second aspect of the present invention, the present invention provides a traffic risk characterization and assessment device for an urban road intersection, comprising the following modules:

[0043] A trajectory classification module is used to classify the trajectories of moving objects in the intersection using a trajectory classification method based on dynamic time normalization and extract the traffic mode;

[0044] The conflict point identification module is used to use the collision time as a key indicator to measure whether there is a traffic conflict between two vehicles, so as to identify the trajectory point where the traffic conflict occurs, that is, the conflict point;

[0045] The conflict point classification module is used to detect the traffic mode to which the conflict point belongs and classify the conflict point according to the traffic mode to which it belongs. Different traffic modes will generate different initial values ​​of risk diffusion to act on the risk diffusion simulation of a single object.

[0046] The risk index construction module is used to integrate the frequency and severity of traffic conflicts and construct the traffic risk index R to quantify the traffic risk at the conflict point;

[0047] The risk diffusion model construction module is used to combine the Gaussian smoke plume diffusion model to construct a traffic risk diffusion model that conforms to the traffic movement mechanism;

[0048] The risk diffusion processing module is used to take the conflict point as the diffusion source, perform traffic risk diffusion processing according to the traffic risk diffusion model, and obtain the traffic risk distribution of the intersection;

[0049] The intersection-level risk assessment module is used to calculate the diffusion area of ​​traffic risk within the intersection, so as to evaluate the traffic risk level at the intersection level;

[0050] The path-level risk assessment module is used to calculate the traffic risk change curve of the traffic trajectory in the intersection based on the traffic risk distribution of the intersection, and use its integral to evaluate the traffic risk level at the path level;

[0051] The mode-level risk assessment module is used to combine the traffic risk level at the path level and calculate the average of the traffic risk levels of all paths in the traffic mode, so as to evaluate the traffic risk level at the traffic mode level.

[0052] According to the third aspect of the present invention, the present invention provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the steps of the method for characterizing and assessing traffic risks at urban road intersections are implemented.

[0053] The technical solution provided by the present invention has the following beneficial effects:

[0054] The smoke plume diffusion model is introduced to simulate the process of propagation and diffusion of traffic risks from conflict points to the neighborhood, characterize the risk distribution of the entire intersection, and serve the needs of high concurrency, low latency, global traffic, and high-precision real-time digital mapping of all elements. Based on the intersection risk distribution, a multi-dimensional hierarchical evaluation system for fine traffic risks at intersections is implemented, mainly including overall risks based on area, path traffic risks based on a single motion trajectory, and traffic mode risks based on multiple trajectories. The quantitative model of global traffic risk at intersections and the multi-dimensional hierarchical evaluation system constructed by the present invention enable risk assessment under single-path or group traffic modes through risk field simulation methods, which is expected to play an important supporting role in future urban risk analysis, and thus provide technical support for the development and construction of smart cars, smart transportation, safe travel, and new smart cities in my country. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] The specific effects of the present invention will be further described below in conjunction with the accompanying drawings and embodiments, in which:

[0056] Figure 1 It is a flow chart of a method for characterizing and evaluating traffic risks at urban road intersections of the present invention;

[0057] Figure 2 It is the intersection map of Guanshan Avenue and Xinyu Road in Wuhan City adopted by the present invention;

[0058] Figure 3 is a trajectory classification result diagram of the present invention;

[0059] Figure 4 The present invention is a distribution diagram of original trajectory points of moving objects and traffic conflict points at the intersection of Guanshan Avenue and Xinyu Road in Wuhan;

[0060] Figure 5 The present invention is a traffic risk distribution diagram of twelve time periods after the risk diffusion at the intersection of Guanshan Avenue and Xinyu Road in Wuhan;

[0061] Figure 6 The traffic risk diffusion area ratio curve diagram of the intersection of Guanshan Avenue and Xinyu Road in Wuhan City of the present invention;

[0062] Figure 7 is the risk distribution curve of the traffic trajectory at the intersection of Guanshan Avenue and Xinyu Road in Wuhan City of the present invention; RL, Mean and t are the traffic risk index of the path, the traffic risk integral of each trajectory point on the path, and the average traffic risk and travel time of each trajectory point on the path, respectively;

[0063] Figure 8 The traffic risk distribution map of each traffic mode at the intersection of Guanshan Avenue and Xinyu Road in Wuhan City of the present invention;

[0064] Fig. 9 It is a structural diagram of a traffic risk characterization and assessment device for an urban road intersection according to the present invention. DETAILED DESCRIPTION

[0065] In order to have a clearer understanding of the technical features, purposes and effects of the present invention, specific embodiments of the present invention are now described in detail with reference to the accompanying drawings.

[0066] Please refer to Figure 1 The present invention provides a method for characterizing and evaluating traffic risks at urban road intersections, comprising the following steps:

[0067] Step S1: using the dynamic time warping (DTW) trajectory classification method to classify the trajectories of moving objects in the intersection and extract the traffic mode; specifically, firstly using the dynamic time warping (DTW) to calculate the similarity between the traffic trajectories in the intersection, setting a suitable threshold to classify the trajectories, the classification result can meet the effect of distinguishing the traffic mode in the intersection, and extract the traffic mode type in the intersection;

[0068] Step S2: using the key indicator - time to collision (TTC) to measure whether the two vehicles have a traffic conflict, so as to identify the trajectory point where the traffic conflict occurs;

[0069] Conflict point identification means: by calculating the collision time (TTC) of all trajectory points in the trajectory, setting a collision time threshold, and the trajectory points with collision time higher than the threshold are conflict points, and they are screened out.

[0070] Step S3: Detect the traffic mode to which the conflict point belongs, and classify the conflict point accordingly;

[0071] Conflict point classification means: based on the extracted internal traffic mode of the intersection, the traffic mode to which all conflict points belong is detected, and the conflict points are classified accordingly.

[0072] Step S4: Using the frequency and severity of traffic conflicts in the traffic conflict technical concept, a risk index R is constructed to quantify the traffic risk at the conflict point;

[0073] Quantifying the traffic risk of conflict points means: first, using the traffic conflict frequency Sum and severity SI in the traffic conflict technical concept, a risk index R is constructed. Then, the traffic risk of all conflict points is calculated based on the risk index R. The traffic risk of all conflict points is calculated based on the risk index R, and the construction formula is as follows:

[0074] R i =0.5×Sum i +0.5×SI i

[0075] In the formula, R i represents the traffic risk value of type i conflict point; Sum i represents the traffic conflict frequency of type i conflict point; SI i Indicates the severity of traffic conflict at type i conflict point.

[0076] Step S5: Combined with the Gaussian smoke plume diffusion model, a traffic risk diffusion model that conforms to the traffic movement mechanism is constructed;

[0077] In order to simulate the diffusion process of traffic risk, the risk diffusion is regarded as pollutants, and the process of pollutants diffusing to the surrounding environment is simulated; in view of the directional characteristics of traffic risk diffusion, the smoke plume diffusion model is introduced to simulate the propagation of risk along the direction of vehicle movement; the smoke plume model uses the horizontal wind direction to drive the diffusion of smoke. We use the directional movement speed of objects along different trajectory paths to simulate the risk diffusion of traffic conflict points to their neighboring areas. Among these paths, the limited risk diffusion path is called the non-risk diffusion path. The smoke plume model initially used the concentration of smoke pollutants as a diffusion index. The present invention uses the traffic risk value as a diffusion index. The diffusion process of the smoke plume model is carried out in three-dimensional space. In the present invention, the diffusion in the vertical direction is not considered, but is limited to two-dimensional plane space. The diffusion direction of traffic risk is set within a 180-degree angle range dominated by the speed direction. In the diffusion of traffic risk, the traffic risk index R is used as the diffusion index, the main diffusion direction is the direction of movement speed, and the 180-degree angle interval of the speed direction is the diffusion interval. The traffic risk diffusion model is constructed as follows:

[0078]

[0079] In the formula, c(x,y) represents the traffic risk value at the position (x,y); δ represents the risk intensity of the diffusion source; μ represents the speed of the source object vehicle; σ y represents the diffusion coefficient.

[0080] Then, the spatially continuous risk field is simulated by interpolating the risk values ​​of the conflict points and the critical risk values ​​obtained after the risk propagates along the path direction using the plume diffusion model. The risk field is constructed by point sets and typical interpolation algorithms that overcome the problem of uneven density distribution. Therefore, we use the plume model and spatial interpolation algorithms to model the risk diffusion phenomenon within the intersection, capture the risk distribution characteristics, and obtain a representation of the intersection risk field.

[0081] Step S6: Taking the conflict point as the diffusion source, traffic risk diffusion processing is performed according to the traffic risk diffusion model to obtain the intersection traffic risk distribution.

[0082] Step S7: Calculate the diffusion area ratio of the traffic risk inside the intersection to evaluate the traffic risk level at the intersection level;

[0083] The diffusion area rate means: taking the risk value P as the threshold, the area greater than or equal to P is the traffic risk area, and the area less than P is the safe area. The traffic risk area is accumulated and the area is calculated to obtain the internal traffic risk diffusion area of ​​the intersection. The calculation formula is as follows:

[0084] S risk =∑s j if s j.risk ≥P

[0085] In the formula, S risk Represents the traffic risk diffusion area inside the intersection; s j represents the area of ​​region j; s j.risk represents the traffic risk value of area j; P represents the traffic risk threshold.

[0086] Step S8: Calculate the traffic risk change curve of the traffic trajectory in the intersection in combination with the traffic risk distribution of the intersection, and use its integral to evaluate the traffic risk level of the path level;

[0087] Traffic risk assessment based on the path level starts from the local perspective of a single moving target. In the process of passing through an intersection, a single object is always facing traffic risks. The traffic risk change curve of the passing trajectory in the intersection is calculated based on the traffic risk distribution of the intersection, and the traffic risk integral along the path is used to evaluate the traffic risk level of a single object when passing through the intersection. The calculation formula is as follows:

[0088]

[0089] Where RL l represents the traffic risk of trajectory l, represents the sum of the traffic risks of all trajectories in trajectory l, t represents the travel time of trajectory l, and α is the adjustment coefficient.

[0090] Step S9: Calculate the average of the traffic risk levels of all paths in the traffic mode in combination with the traffic risk levels of the paths, and use this to evaluate the traffic risk level of the traffic mode level.

[0091] The actual travel path set represents the trajectory selection under the specific traffic mode of the intersection. The mode-level traffic risk assessment aims to quantitatively assess the risk under the specific traffic mode of the intersection. Evaluating the traffic risk level of the driving trajectory set can reflect the traffic risk level of the turning mode. In the present invention, the mean of the traffic risk level of the trajectories included in the same type of traffic mode is used to evaluate the traffic risk of the traffic mode or a certain turning mode. The calculation formula is as follows:

[0092]

[0093] In the formula, RM m represents the traffic risk of traffic mode m, represents the sum of traffic risks of all trajectories in mode m.

[0094] In order to more clearly illustrate the concept of the present invention, the following is a further description of the intersection of Guanshan Avenue and Xinyu Road in Wuhan as an example. Figure 2 , Figure 2 is a map of the intersection of Guanshan Avenue and Xinyu Road in Wuhan City adopted by the present invention, wherein Figure 2 (a) is a real picture of the study area; Figure 2 (b) is the satellite map of the study area. Figure 2 (c) is the street map of the study area.

[0095] The method of the present invention is used to characterize and evaluate traffic risks at urban road intersections. The specific steps are as follows:

[0096] Step S1: Use the dynamic time warping (DTW) trajectory classification method to classify the trajectories of moving objects in the intersection and extract the traffic mode. Specifically, first use dynamic time warping (DTW) to calculate the similarity between the traffic trajectories in the intersection, set a suitable threshold to classify the trajectories, and the classification result can distinguish the traffic mode in the intersection and extract the traffic mode type in the intersection. The result is as follows Figure 3 As shown, Figure 3 (a)-(o) correspond to 15 traffic modes. Table 1 shows the number of trajectories in the 15 traffic modes.

[0097] Table 1: Number of trajectories for each traffic mode

[0098]

[0099]

[0100] In Table 1, Model represents the traffic mode, and TrN represents the number of trajectories.

[0101] Step S2: Using the key indicator - time to collision (TTC) to measure whether the two vehicles have a traffic conflict, so as to identify the trajectory point where the traffic conflict occurs;

[0102] Conflict point identification means: by calculating the collision time (TTC) of all track points in the trajectory, setting the collision time threshold to 1.5s, the track points with collision time higher than the threshold are considered conflict points and are screened out. Figure 4 As shown, Figure 4 (a) corresponds to the original trajectory point, Figure 4 (b) Corresponding traffic conflict points.

[0103] Step S3: Detect the traffic mode to which the conflict point belongs, and classify the conflict point accordingly;

[0104] Conflict point classification means: according to the extracted traffic mode inside the intersection, the traffic mode to which all conflict points belong is detected, and the conflict points are classified accordingly. The results are shown in Table 2:

[0105] Table 2 Number of trajectories, number of trajectory points, and number of conflict points in each mode

[0106]

[0107] Step S4: Using the frequency and severity of traffic conflicts in the traffic conflict technical concept, a risk index R is constructed to quantify the traffic risk at the conflict point;

[0108] Quantifying the traffic risk of conflict points means: first, using the traffic conflict frequency Sum and severity SI in the traffic conflict technical concept, a risk index R is constructed. Then, the traffic risk of all conflict points is calculated based on the risk index R.

[0109] Step 5) Combined with the Gaussian smoke plume diffusion model, a traffic risk diffusion model that conforms to the traffic movement mechanism is constructed;

[0110] The traffic risk diffusion model refers to: based on the Gaussian plume diffusion model, the risk diffusion ratio is used as the pollutant, and the process of pollutants diffusing to the surrounding area is used to simulate the traffic risk diffusion process. In the traffic risk diffusion, the traffic risk index R is used as the diffusion index, the main diffusion direction is the direction of movement speed, and the 180-degree angle interval of the speed direction is the diffusion interval.

[0111] Step 6) Taking the conflict point as the diffusion source, the traffic risk diffusion model is used to perform traffic risk diffusion processing to obtain the traffic risk distribution of the intersection. The maximum traffic risk results for each time period are as follows: Figure 5 As shown in Table 3, the maximum traffic risk at the intersection in each time period.

[0112] Table 3: Maximum traffic risks at intersections during 12 time periods

[0113]

[0114] Step 7) Calculate the diffusion area of ​​traffic risk inside the intersection to evaluate the traffic risk level at the intersection level;

[0115] The diffusion area rate means: taking the risk value P as the threshold, the area greater than or equal to P is the traffic risk area, and the area less than P is the safe area. The traffic risk area is accumulated and the area is calculated to obtain the traffic risk diffusion area inside the intersection. The result is as follows Figure 6 shown.

[0116] Step 8) Calculate the traffic risk change curve of the traffic trajectory in the intersection in combination with the traffic risk distribution of the intersection, and use its integral to evaluate the traffic risk level at the path level;

[0117] The path-level traffic risk level refers to: calculating the traffic risk change curve of the passage trajectory based on the traffic risk distribution of the intersection, and using the traffic risk integral along the path to evaluate the traffic risk of a single object when passing through the intersection. Figure 7 As shown in Table 4, the traffic risk status of each trajectory is described.

[0118] Table 4: Traffic risk status of trajectory

[0119]

[0120] In Table 4, RL represents the traffic risk value of the trajectory. represents the sum of the traffic risk values ​​of all trajectory points in the trajectory, Mean represents the mean of the traffic risk values ​​of all trajectory points in the trajectory, and t represents the travel time of the trajectory.

[0121] Step 9) Combined with the traffic risk level of the path, the mean of the traffic risk level of all paths in the traffic mode is calculated to evaluate the traffic risk level at the traffic mode level.

[0122] The traffic risk level at the mode level refers to: using the average traffic risk of the same type of traffic mode including the trajectory to evaluate the traffic risk of the traffic mode. Figure 8 As shown in Table 5, the traffic risks of each travel mode are described.

[0123] Table 5: Traffic risks of each mode of travel

[0124]

[0125]

[0126] In Table 5, Model represents the traffic mode, and RM represents the traffic risk of the traffic mode.

[0127] In addition, in order to implement the above-mentioned traffic risk characterization and assessment method, this embodiment also provides a traffic risk characterization and assessment device for urban road intersections. Fig. 9 As shown, the device includes the following modules:

[0128] A trajectory classification module 1 is used to classify the trajectories of moving objects in the intersection using a trajectory classification method based on dynamic time normalization to extract the traffic mode;

[0129] The conflict point identification module 2 is used to use the collision time as a key indicator to measure whether the two vehicles have a traffic conflict, so as to identify the trajectory point where the traffic conflict occurs, that is, the conflict point;

[0130] The conflict point classification module 3 is used to detect the traffic mode to which the conflict point belongs and classify the conflict point according to the traffic mode to which it belongs. Different traffic modes will generate different initial values ​​of risk diffusion to act on the risk diffusion simulation of a single object.

[0131] The risk index construction module 4 is used to integrate the traffic conflict frequency and severity to construct the traffic risk index R, so as to quantify the traffic risk at the conflict point;

[0132] The risk diffusion model construction module 5 is used to combine the Gaussian smoke plume diffusion model to construct a traffic risk diffusion model that conforms to the traffic movement mechanism;

[0133] The risk diffusion processing module 6 is used to take the conflict point as the diffusion source, perform traffic risk diffusion processing according to the traffic risk diffusion model, and obtain the traffic risk distribution of the intersection;

[0134] The intersection level risk assessment module 7 is used to calculate the diffusion area of ​​the traffic risk inside the intersection, so as to evaluate the traffic risk level at the intersection level;

[0135] A path level risk assessment module 8 is used to calculate the traffic risk change curve of the traffic trajectory in the intersection in combination with the traffic risk distribution of the intersection, and use its integral to evaluate the traffic risk level of the path level;

[0136] The mode-level risk assessment module 9 is used to calculate the average of the traffic risk levels of all paths in the traffic mode in combination with the traffic risk level at the path level, so as to assess the traffic risk level at the traffic mode level.

[0137] In addition, this embodiment also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the steps of the method for characterizing and evaluating traffic risks at urban road intersections are implemented.

[0138] The embodiment of the present invention proposes a method, device and equipment for characterizing and evaluating traffic risks at urban road intersections. By establishing a multidimensional risk quantification model, the temporal and spatial dynamic risk distribution of intersections is revealed, providing a quantitative reference for risk decision-making. First, a traffic risk index is constructed in combination with the frequency of traffic conflicts and the severity of conflicts to quantitatively characterize the traffic risks of traffic conflict points in intersections. Then, the interactive propagation of traffic risks at conflict points in the spatial neighborhood is simulated, and a plume diffusion model is introduced to simulate the propagation and diffusion process of traffic risks at conflict points to the neighborhood, characterizing the risk distribution of the entire intersection, and serving the needs of high concurrency, low latency, global traffic, and high-precision real-time digital mapping of all elements. Finally, based on the intersection risk distribution, a multi-dimensional hierarchical evaluation system for refined traffic risks at intersections is implemented, mainly including overall risk based on area, path traffic risk based on a single motion trajectory, and traffic mode risk based on multiple trajectories. The intersection-wide traffic risk quantification model and multi-dimensional hierarchical evaluation system constructed in this embodiment enable risk assessment under single-path or group traffic modes through risk field simulation methods, and are expected to play an important supporting role in future urban risk analysis, thereby providing technical support for the development and construction of my country's smart cars, smart transportation, safe travel and new smart cities.

[0139] It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or system including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or system. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the existence of other identical elements in the process, method, article or system including the element.

[0140] The serial numbers of the embodiments of the present invention are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments. In a unit claim that lists several means, several of these means may be embodied by the same hardware item. The use of the words first, second, and third, etc. does not indicate any order and these words may be interpreted as identifiers.

[0141] The above are only preferred embodiments of the present invention, and are not intended to limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made using the contents of the present invention specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A method for characterizing and assessing traffic risks at urban road intersections, characterized in that: The following steps are involved: The trajectory classification method of dynamic time normalization is used to classify the trajectories of moving objects in the intersection and extract the traffic mode; The collision time is used as a key indicator to measure whether there is a traffic conflict between two vehicles, so as to identify the trajectory point where the traffic conflict occurs, namely the conflict point; Detect the traffic mode to which the conflict point belongs, and classify the conflict point according to the traffic mode to which it belongs. Different traffic modes will generate different initial values ​​of risk diffusion to act on the risk diffusion simulation of a single object; By integrating the frequency and severity of traffic conflicts, a traffic risk index R is constructed to quantify the traffic risk at the conflict point. Combined with the Gaussian smoke plume diffusion model, a traffic risk diffusion model that conforms to the traffic movement mechanism is constructed; Taking the conflict point as the diffusion source, the traffic risk diffusion model is used to perform traffic risk diffusion processing to obtain the traffic risk distribution of the intersection; Calculate the diffusion area of ​​traffic risk inside the intersection to evaluate the traffic risk level at the intersection level; The traffic risk change curve of the traffic trajectory in the intersection is calculated based on the traffic risk distribution of the intersection, and its integral is used to evaluate the traffic risk level at the path level; Combined with the traffic risk level at the path level, the mean of the traffic risk levels of all paths in the traffic mode is calculated to evaluate the traffic risk level at the traffic mode level.

2. The method for characterizing and evaluating traffic risks at urban road intersections according to claim 1 is characterized in that: The step of using the trajectory classification method of dynamic time normalization to classify the trajectories of moving objects in the intersection and extracting the traffic mode includes: Dynamic time normalization is used to calculate the similarity between the traffic trajectories inside the intersection, and an appropriate threshold is set to classify the trajectories. The classification result is sufficient to distinguish the traffic modes inside the intersection and extract the types of traffic modes in the intersection, which is used to assign different risk diffusion initial values ​​to different conflict points.

3. The traffic risk characterization and assessment method for urban road intersections according to claim 1 is characterized in that: The step of using the collision time as a key indicator to measure whether a traffic conflict occurs between two vehicles, so as to identify the trajectory point where the traffic conflict occurs, includes: By calculating the collision time of all trajectory points in the trajectory and setting the collision time threshold, trajectory points with collision time higher than the threshold are considered conflict points and are screened out.

4. The method for characterizing and evaluating traffic risks at urban road intersections according to claim 1 is characterized in that: The steps of integrating the traffic conflict frequency and severity to construct a traffic risk index R to quantify the traffic risk of the conflict point include: With the help of the traffic conflict frequency Sum and severity SI in the traffic conflict technical concept, a risk index R is constructed; The traffic risk of various conflict points is calculated based on the risk index R, and the formula is as follows: R i =0.5×Sum i +0.5×SI i In the formula, R i represents the traffic risk value of type i conflict point; Sum i represents the traffic conflict frequency of type i conflict point; SI i Indicates the severity of traffic conflict at type i conflict point.

5. The method for characterizing and evaluating traffic risks at urban road intersections according to claim 1 is characterized in that: The step of combining the Gaussian smoke plume diffusion model to construct a traffic risk diffusion model that conforms to the traffic movement mechanism includes: According to the Gaussian plume diffusion model, the risk diffusion ratio is used as the pollutant, and the process of pollutant diffusion to the surrounding area is used to simulate the traffic risk diffusion process; The traffic risk index R is used as the diffusion index, the main diffusion direction is the movement speed direction, and the 180-degree angle interval in the speed direction is used as the diffusion interval. The traffic risk diffusion model is constructed as follows: In the formula, c(x,y) represents the traffic risk value at the position (x,y); δ represents the risk intensity of the diffusion source; μ represents the speed of the source object vehicle; σ y represents the diffusion coefficient.

6. The method for characterizing and evaluating traffic risks at urban road intersections according to claim 1 is characterized in that: The step of calculating the diffusion area rate of the internal traffic risk of the intersection comprises: Taking the risk value P as the threshold, the area greater than or equal to P is the traffic risk area, and the area less than P is the safe area. The traffic risk area is accumulated and the area is calculated to obtain the traffic risk diffusion area inside the intersection. The calculation formula is as follows: S risk =∑s j if s j.risk ≥P In the formula, S risk Represents the traffic risk diffusion area inside the intersection; s j represents the area of ​​region j; s j.risk represents the traffic risk value of area j; P represents the traffic risk threshold.

7. The method for characterizing and evaluating traffic risks at urban road intersections according to claim 1 is characterized in that: The step of calculating the traffic risk change curve of the traffic trajectory in the intersection in combination with the traffic risk distribution of the intersection, and using the integral thereof to evaluate the traffic risk level of the path level includes: The traffic risk change curve of the traffic trajectory in the intersection is calculated according to the traffic risk distribution of the intersection. The traffic risk integral along the path is used to evaluate the traffic risk level of a single object when passing through the intersection. The calculation formula is as follows: Where RL l represents the traffic risk of trajectory l, represents the sum of the traffic risks of all trajectories in trajectory l, t represents the travel time of trajectory l, and α is the adjustment coefficient.

8. The method for characterizing and evaluating traffic risks at urban road intersections according to claim 7 is characterized in that: The step of calculating the average of the traffic risk levels of all paths in the traffic mode in combination with the traffic risk level at the path level to evaluate the traffic risk level at the traffic mode level includes: The traffic risk of a traffic mode or a certain turning mode is evaluated by using the mean of the traffic risk levels of the same type of traffic modes including trajectories. The calculation formula is as follows: In the formula, RM m represents the traffic risk of traffic mode m, represents the sum of traffic risks of all trajectories in mode m.

9. A traffic risk characterization and assessment device for urban road intersections, characterized in that: Includes the following modules: A trajectory classification module is used to classify the trajectories of moving objects in the intersection using a trajectory classification method based on dynamic time normalization and extract the traffic mode; The conflict point identification module is used to use the collision time as a key indicator to measure whether there is a traffic conflict between two vehicles, so as to identify the trajectory point where the traffic conflict occurs, that is, the conflict point; The conflict point classification module is used to detect the traffic mode to which the conflict point belongs and classify the conflict point according to the traffic mode to which it belongs. Different traffic modes will generate different initial values ​​of risk diffusion to act on the risk diffusion simulation of a single object. The risk index construction module is used to integrate the frequency and severity of traffic conflicts and construct the traffic risk index R to quantify the traffic risk at the conflict point; The risk diffusion model construction module is used to combine the Gaussian smoke plume diffusion model to construct a traffic risk diffusion model that conforms to the traffic movement mechanism; The risk diffusion processing module is used to take the conflict point as the diffusion source, perform traffic risk diffusion processing according to the traffic risk diffusion model, and obtain the traffic risk distribution of the intersection; The intersection-level risk assessment module is used to calculate the diffusion area of ​​traffic risk within the intersection, so as to evaluate the traffic risk level at the intersection level; The path-level risk assessment module is used to calculate the traffic risk change curve of the traffic trajectory in the intersection based on the traffic risk distribution of the intersection, and use its integral to evaluate the traffic risk level at the path level; The mode-level risk assessment module is used to combine the traffic risk level at the path level and calculate the average of the traffic risk levels of all paths in the traffic mode, so as to evaluate the traffic risk level at the traffic mode level.

10. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, the steps of the method for characterizing and evaluating traffic risks at urban road intersections as described in any one of claims 1 to 8 are implemented.

Images