Expressway risk assessment method, device and equipment and storage medium

Abstract

The invention discloses an expressway risk assessment method, device and equipment and a storage medium, and relates to the technical field of traffic safety, and the method comprises the steps: collecting vehicle track data and road environment data corresponding to a target road section in an expressway confluence area; based on the vehicle trajectory data and the road environment data, constructing multi-element coupling features of the highway confluence area; performing scene clustering on the multi-element coupling features through a preset confluence area risk scene clustering model to obtain a scene clustering result; and determining a high-risk scene in the highway confluence area based on a scene clustering result. According to the method, the multi-element coupling features can be constructed based on the vehicle trajectory data and the road environment data, scene clustering is performed on the multi-element coupling features, and then the high-risk scene in the highway confluence area is determined based on the scene clustering result. The problem that a risk assessment method in the prior art is difficult to accurately identify a high-risk state in a driving process and cannot timely perform high-risk scene early warning is solved.

Description

Technical Field

[0001] This application relates to the technical field of traffic safety, and particularly to a highway risk assessment method, device, equipment, and storage medium. Background Art

[0002] Research shows that the highway merging area is one of the high-incidence areas of traffic accidents in China. Among them, traffic accidents in the highway merging area often result from the coupling imbalance of multiple elements of human-vehicle-road-environment in a high-dynamic scenario. Especially in the coupling of factors such as fatigue driving, bad weather, unclear traffic guidance, and night driving, the complex interaction further exacerbates the system instability and forms a high-risk scenario. With the rapid development of technology, intelligent vehicles have gradually been put into use on highways. However, at present, there are still certain technical bottlenecks when intelligent vehicles deal with high-risk scenarios in the merging area.

[0003] Currently, traditional risk assessment methods usually conduct risk assessment based on the mapping relationship between environmental variables and traffic safety indicators. However, due to the large uncertainty of environmental changes, the reliability of using this method for risk assessment is not high, making it difficult to accurately identify high-risk states during driving, and thus unable to timely give early warnings for high-risk scenarios. Summary of the Invention

[0004] The main purpose of this application is to provide a highway risk assessment method, device, equipment, and storage medium, aiming to solve the technical problem that the existing risk assessment method is difficult to accurately identify high-risk states during driving and unable to timely give early warnings for high-risk scenarios.

[0005] To achieve the above purpose, this application proposes a highway risk assessment method, and the method includes:

[0006] Collect vehicle trajectory data and road environment data corresponding to the target section in the highway merging area;

[0007] Construct multi-element coupling characteristics of the highway merging area based on the vehicle trajectory data and the road environment data;

[0008] Perform scenario clustering on the multi-element coupling characteristics through a preset merging area risk scenario clustering model to obtain a scenario clustering result;

[0009] Determine high-risk scenarios in the highway merging area based on the scenario clustering result.

[0010] In an embodiment, before the step of performing scenario clustering on the multi-element coupling characteristics through a preset merging area risk scenario clustering model to obtain a scenario clustering result, it further includes:

[0011] Determine the potential field intensity generated by several elements in the highway merging area based on the multi-element coupling characteristics through a preset driving risk field model for the merging area, where the potential field intensity is the potential field intensity of each element on the merging vehicle in the highway merging area;

[0012] Determine the driving risk situation value corresponding to the highway merging area according to the potential field intensity and the potential field superposition theory;

[0013] The step of determining the high-risk scenarios in the highway merging area based on the scenario clustering result includes:

[0014] Determine the high-risk scenarios in the highway merging area based on the scenario clustering result and the driving risk situation value.

[0015] In one embodiment, the potential field intensity includes: road boundary potential field intensity, lane line potential field intensity, and construction area potential field intensity; the step of determining the potential field intensity generated by several elements in the highway merging area based on the multi-element coupling characteristics through a preset driving risk field model for the merging area includes:

[0016] Through a preset driving risk field model for the merging area, determine the lane line distance between the vehicle driving in the highway merging area and the center line of the lane, the current speed of the vehicle, the road driving speed threshold, as well as the boundary distance, safety distance, and direction angle between the vehicle and the construction area boundary according to the multi-element coupling characteristics;

[0017] Based on the lane line distance, the current speed of the vehicle, and the road driving speed threshold, determine the road boundary potential field intensity corresponding to the road boundary in the highway merging area and the lane line potential field intensity corresponding to the lane line;

[0018] Based on the boundary distance, the safety distance, and the direction angle, determine the construction area potential field intensity corresponding to the construction area in the highway merging area;

[0019] The step of determining the driving risk situation value corresponding to the highway merging area according to the potential field intensity and the potential field superposition theory includes:

[0020] Determine the driving risk situation value corresponding to the highway merging area through the potential field superposition theory based on the road boundary potential field intensity, the lane line potential field intensity, and the construction area potential field intensity.

[0021] In one embodiment, the potential field intensity further includes: vehicle interaction potential field intensity; the step of determining the potential field intensity generated by several elements in the highway merging area based on the multi-element coupling characteristics through a preset driving risk field model for the merging area includes:

[0022] Based on the preset confluence area driving risk field model, determine the steering angle, vehicle type of the merging vehicle in the highway confluence area, as well as the relative speed and relative distance from surrounding vehicles based on the multi-factor coupling characteristics;

[0023] Based on the steering angle, the vehicle type, the relative speed and the relative distance, determine the vehicle interaction potential field intensity generated by vehicle interaction in the highway confluence area;

[0024] The step of determining the driving risk situation value corresponding to the highway confluence area based on the road boundary potential field intensity, the lane line potential field intensity and the construction area potential field intensity through the potential field superposition theory includes:

[0025] Determine the driving risk situation value corresponding to the highway confluence area based on the road boundary potential field intensity, the lane line potential field intensity, the construction area potential field intensity and the vehicle interaction potential field intensity through the potential field superposition theory.

[0026] In one embodiment, the step of determining the driving risk situation value corresponding to the highway confluence area based on the road boundary potential field intensity, the lane line potential field intensity, the construction area potential field intensity and the vehicle interaction potential field intensity through the potential field superposition theory includes:

[0027] Superpose the road boundary potential field intensity, the lane line potential field intensity, the construction area potential field intensity and the vehicle interaction potential field intensity through the potential field superposition theory to obtain the synthetic potential field intensity;

[0028] Based on the preset reference risk value, the preset risk growth coefficient and the synthetic potential field intensity, determine the driving risk situation value corresponding to the highway confluence area.

[0029] In one embodiment, the step of performing scenario clustering on the multi-factor coupling characteristics through the preset confluence area risk scenario clustering model to obtain a scenario clustering result includes:

[0030] Perform low-dimensional mapping on the multi-factor coupling characteristics through the preset confluence area risk scenario clustering model to obtain low-dimensional multi-factor coupling characteristics;

[0031] Restore the low-dimensional multi-factor coupling characteristics to the original feature space to obtain the restored multi-factor coupling characteristics;

[0032] Adopt a soft assignment method to perform scenario clustering on the restored multi-factor coupling characteristics to obtain a scenario clustering result.

[0033] In one embodiment, the step of determining the high-risk scenarios in the highway confluence area based on the scenario clustering result and the driving risk situation value includes:

[0034] Determine the average risk level corresponding to each clustering scenario in the highway merging area based on the scenario clustering result and the driving risk situation value;

[0035] Determine the optimal dynamic risk threshold according to the average risk level;

[0036] Determine the high-risk scenarios in the highway merging area based on the optimal dynamic risk threshold.

[0037] In addition, to achieve the above object, the present application also proposes a highway risk assessment device, and the device includes:

[0038] A data acquisition module, configured to acquire vehicle trajectory data and road environment data corresponding to a target section in the highway merging area;

[0039] A feature construction module, configured to construct multi-factor coupling features of the highway merging area based on the vehicle trajectory data and the road environment data;

[0040] A scenario clustering module, configured to perform scenario clustering on the multi-factor coupling features through a preset merging area risk scenario clustering model to obtain a scenario clustering result;

[0041] A risk assessment module, configured to determine the high-risk scenarios in the highway merging area based on the scenario clustering result.

[0042] In addition, to achieve the above object, the present application also proposes a highway risk assessment device, and the device includes: a memory, a processor, and a computer program stored on the memory and executable on the processor, and the computer program is configured to implement the steps of the highway risk assessment method as described above.

[0043] In addition, to achieve the above object, the present application also proposes a storage medium, the storage medium is a computer-readable storage medium, and a computer program is stored on the storage medium, and when the computer program is executed by a processor, the steps of the highway risk assessment method as described above are implemented.

[0044] The present application provides a method for risk assessment of expressways. The present application discloses collecting vehicle trajectory data and road environment data corresponding to a target section in an expressway merging area; constructing multi-factor coupling characteristics of the expressway merging area based on the vehicle trajectory data and the road environment data; performing scenario clustering on the multi-factor coupling characteristics through a preset merging area risk scenario clustering model to obtain a scenario clustering result; determining high-risk scenarios in the expressway merging area based on the scenario clustering result. Compared with the existing risk assessment methods which usually perform risk assessment based on the mapping relationship between environmental variables and traffic safety indicators, and the uncertainty of environmental changes is relatively large, resulting in low reliability of risk assessment. Since the present invention can construct multi-factor coupling characteristics based on the vehicle trajectory data and the road environment data corresponding to a target section in the expressway merging area, perform scenario clustering on the multi-factor coupling characteristics, and then determine high-risk scenarios in the expressway merging area based on the scenario clustering result, thus solving the technical problems that the existing risk assessment methods are difficult to accurately identify high-risk states during driving and cannot give early warnings for high-risk scenarios in a timely manner. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] The accompanying drawings herein are incorporated into and constitute a part of this specification, showing embodiments consistent with the present application, and are used together with the specification to explain the principles of the present application.

[0046] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the accompanying drawings required for use in the description of the embodiments or the prior art. Obviously, for those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0047] Figure 1 It is a schematic flowchart provided for Embodiment 1 of the expressway risk assessment method of the present application;

[0048] Figure 2 It is a schematic flowchart provided for Embodiment 2 of the expressway risk assessment method of the present application;

[0049] Figure 3 It is a schematic flowchart provided for Embodiment 3 of the expressway risk assessment method of the present application;

[0050] Figure 4 It is a schematic diagram of the module structure of the expressway risk assessment device according to the embodiment of the present application;

[0051] Figure 5 It is a schematic diagram of the device structure of the hardware operating environment involved in the expressway risk assessment method according to the embodiment of the present application.

[0052] The realization of the object, functional features and advantages of the present application will be further described with reference to the embodiments and the accompanying drawings. Specific Embodiments

[0053] It should be understood that the specific embodiments described herein are only used to explain the technical solutions of the present application and are not used to limit the present application.

[0054] For a better understanding of the technical solutions of the present application, the following will be described in detail in conjunction with the accompanying drawings of the specification and specific embodiments.

[0055] The main solution of the embodiments of the present application is: collect vehicle trajectory data and road environment data corresponding to a target section in the highway merge area; construct multi-element coupling features of the highway merge area based on the vehicle trajectory data and road environment data; perform scenario clustering on the multi-element coupling features through a preset merge area risk scenario clustering model to obtain a scenario clustering result; determine high-risk scenarios in the highway merge area based on the scenario clustering result.

[0056] Since the risk assessment methods in the prior art usually perform risk assessment based on the mapping relationship between environmental variables and traffic safety indicators, and the uncertainty of environmental changes is relatively large, the reliability of risk assessment is not high.

[0057] The present application provides a solution, which can construct multi-element coupling features based on vehicle trajectory data and road environment data corresponding to a target section in the highway merge area, perform scenario clustering on the multi-element coupling features, and then determine high-risk scenarios in the highway merge area based on the scenario clustering result, thereby solving the technical problems that the risk assessment methods in the prior art are difficult to accurately identify high-risk states during driving and cannot give early warnings for high-risk scenarios in a timely manner.

[0058] It should be noted that the execution subject of this embodiment can be a computing service device with data processing, network communication, and program running functions, such as a tablet computer, a personal computer, a mobile phone, etc., or an electronic device, a highway risk assessment device, etc. that can implement the above functions. Hereinafter, taking a highway risk assessment device as an example (hereinafter referred to as the device), this embodiment and the following embodiments will be described.

[0059] Based on this, the embodiments of the present application provide a highway risk assessment method, referring to Figure 1 , Figure 1 is a schematic flowchart of the first embodiment of the highway risk assessment method of the present application.

[0060] In this embodiment, the highway risk assessment method includes steps S10 to S40:

[0061] Step S10: Collect vehicle trajectory data and road environment data corresponding to a target section in the highway merge area.

[0062] It should be noted that the above target section can be a typical section in the confluence area of an expressway, such as a ramp, a construction section, etc., and this embodiment does not limit this.

[0063] It can be understood that the above vehicle trajectory data can be the driving trajectory data of vehicles entering the target section in the confluence area of an expressway. For example, the position, speed, acceleration, steering angle, etc. of the vehicle. The above road environment data can be the road environment data of the target section in the confluence area of an expressway. For example, the road gradient, visibility, friction, number of lanes, road boundary, construction area, etc. of this section, and this embodiment does not limit this.

[0064] In this embodiment, first, the driving information data and road environment information data of vehicles entering the target section in the confluence area of an expressway can be collected by drones in the target section of the confluence area of an expressway, including the position, speed, and steering angle information of each vehicle, and the position and speed information of surrounding vehicles, and the road boundary, lane line information, speed limit, gradient, friction coefficient, visibility, etc. can be determined through cameras and road section detectors to obtain vehicle trajectory data and road environment data.

[0065] Step S20: Construct the multi-factor coupling characteristics of the expressway confluence area based on the vehicle trajectory data and the road environment data.

[0066] It should be noted that the above multi-factor coupling characteristics can be characteristics that comprehensively integrate multi-dimensional information of people, vehicles, roads, and the environment. In practical applications, the interaction and influence relationship between each factor can be characterized by the multi-factor coupling characteristics.

[0067] In this embodiment, the device can construct the above multi-factor coupling characteristics based on the dynamic characteristics of vehicles driving in the target section of the expressway confluence area, such as speed, acceleration, steering angle, etc., road environment factors such as road gradient, friction, number of lanes, road boundary, construction area, etc., and the type of vehicle, driver's driving style, and fatigue level.

[0068] Step S30: Perform scenario clustering on the multi-factor coupling characteristics through a preset confluence area risk scenario clustering model to obtain a scenario clustering result.

[0069] It should be noted that the above-mentioned preset risk scenario clustering model for the merging area can be a model used to cluster multi-factor coupling features to divide the risk scenarios in the merging area into different types. Correspondingly, the above-mentioned scenario clustering result can be used to represent the categories corresponding to each risk scenario in the highway merging area. In this embodiment, the device can use the soft assignment method to cluster the multi-factor coupling features to determine the probability that each risk scenario in the highway merging area belongs to a certain scenario cluster, so as to obtain the scenario clustering result. Among them, the soft assignment method can be a clustering method based on probability or fuzzy theory. Different from the traditional "hard assignment" (such as K-means), it allows data points to belong to multiple clusters with different probabilities or membership degrees at the same time.

[0070] Step S40: Determine the high-risk scenarios in the highway merging area based on the scenario clustering result.

[0071] In practical applications, after determining the probability that each risk scenario in the highway merging area belongs to a certain scenario cluster through the scenario clustering result, the average risk level of each scenario cluster in the highway can be calculated, and then the maximum inter-class variance method is introduced to determine the high-risk scenarios in the highway merging area based on this average risk level. Among them, the maximum inter-class variance method is a classic algorithm for image segmentation, aiming to automatically determine the global threshold of the image by maximizing the inter-class variance and divide the image into two categories: foreground and background. In this embodiment, the threshold corresponding to the maximum inter-class variance can be selected as the optimal dynamic risk threshold, and all clustering categories exceeding this threshold are screened as high-risk scenarios, that is, the risk scenarios with an average risk level exceeding the optimal dynamic risk threshold are determined as high-risk scenarios.

[0072] This embodiment provides a highway risk assessment method, which discloses collecting vehicle trajectory data and road environment data corresponding to a target section in the highway merging area; constructing multi-factor coupling features of the highway merging area based on the vehicle trajectory data and the road environment data; performing scenario clustering on the multi-factor coupling features through a preset risk scenario clustering model for the merging area to obtain a scenario clustering result; determining the high-risk scenarios in the highway merging area based on the scenario clustering result; compared with the existing risk assessment methods, which usually perform risk assessment based on the mapping relationship between environmental variables and traffic safety indicators, and the uncertainty of environmental changes is relatively large, resulting in low reliability of risk assessment. Since this embodiment can construct multi-factor coupling features based on the vehicle trajectory data and the road environment data corresponding to the target section in the highway merging area, perform scenario clustering on the multi-factor coupling features, and then determine the high-risk scenarios in the highway merging area based on the scenario clustering result, thus solving the technical problems that the existing risk assessment methods are difficult to accurately identify the high-risk states during driving and cannot give early warnings for high-risk scenarios in a timely manner.

[0073] Based on the first embodiment of this application, in the second embodiment of this application, the same or similar content as in the above-mentioned first embodiment can be referred to the above introduction and will not be repeated hereinafter. On this basis, please refer to Figure 2 , Figure 2 which is a schematic flow chart provided for the second embodiment of the highway risk assessment method of this application.

[0074] In this embodiment, before step S30, the method further includes steps S201 to S202:

[0075] Step S201: Based on the preset driving risk field model of the merging area, determine the potential field intensity generated by several elements in the highway merging area according to the multi-element coupling characteristics, and the potential field intensity is the potential field intensity of each element on the merging vehicle in the highway merging area.

[0076] It should be noted that the above potential field intensity can be a value used to characterize the influence of each element in the highway merging area on vehicle driving. In this embodiment, the potential field intensity includes: road boundary potential field intensity, lane line potential field intensity, and construction area potential field intensity. Among them, the road boundary potential field intensity can be the binding force or repulsive force generated by the physical boundaries on both sides of the road (such as guardrails, curbstones or virtual boundaries) on the vehicle; the lane line potential field intensity can be the guiding force or binding force generated by the lane lines (such as solid lines, dotted lines) on the vehicle; the construction area potential field intensity can be the interference force or binding force generated by the road construction area on the vehicle.

[0077] Step S202: Determine the driving risk situation value corresponding to the highway merging area according to the potential field intensity and the potential field superposition theory.

[0078] It can be understood that the potential field superposition theory is that when multiple potential fields (such as gravitational field, electric field, magnetic potential field, etc.) act on a certain space area at the same time, the total potential field intensity at this point is equal to the algebraic sum of the potential field intensities generated by each independent potential field at this point. This theory is based on the superposition property of linear systems, that is, the effects of multiple independent sources can be simply superimposed without interfering with each other.

[0079] It should be understood that the above driving risk situation value can be a value used to characterize the risk degree when the vehicle is driving in the highway merging area. In this embodiment, first, the potential field intensities generated by each element in the highway merging area can be superimposed based on the potential field superposition theory, and the driving risk situation value can be calculated based on the potential field intensity obtained after the superposition.

[0080] Further, the potential field intensity includes: road boundary potential field intensity, lane line potential field intensity, and construction area potential field intensity; the step S201 includes: determining, according to the multi-factor coupling characteristics through a preset driving risk field model for the highway merge area, the lane line distance between the vehicle traveling in the highway merge area and the lane center line, the current vehicle speed, the road driving speed threshold, as well as the boundary distance, safety distance, and direction angle between the vehicle and the construction area boundary; determining the road boundary potential field intensity corresponding to the road boundary in the highway merge area and the lane line potential field intensity corresponding to the lane line based on the lane line distance, the current vehicle speed, and the road driving speed threshold; and determining the construction area potential field intensity corresponding to the construction area in the highway merge area based on the boundary distance, the safety distance, and the direction angle.

[0081] In practical applications, the road boundary potential field intensity E road and the lane line potential field intensity E Lane can be calculated according to data such as the distance between the vehicle and the boundary or lane line, slope, friction, visibility, vehicle speed, and road speed limit. The potential field intensity E work of the construction area on the vehicle can be calculated according to data such as the distance between the vehicle and the construction area boundary, safety distance, and direction angle. The specific calculation formula can be:

[0082]

[0083] In the formula, K r , K l is the weight coefficient, λ r , λ l is the attenuation coefficient, d l is the distance between the vehicle and the lane center line, α s is the slope influence factor, μ is the friction coefficient, θ is the road slope angle, k υ is the speed adjustment coefficient, V is the current vehicle speed, V max is the road speed limit, that is, the above-mentioned road driving speed threshold, d x is the distance between the vehicle and the construction area boundary, d0 is the safety distance, and σ is the direction angle.

[0084] Correspondingly, the step S202 includes:

[0085] Step S202a: Determining the driving risk situation value corresponding to the highway merge area based on the road boundary potential field intensity, the lane line potential field intensity, and the construction area potential field intensity through the potential field superposition theory.

[0086] Further, the potential field strength further includes: vehicle interaction potential field strength; the step S201 includes: determining the steering angle, vehicle type, relative speed, and relative distance from surrounding vehicles of the merging vehicle in the highway merge area based on the multi-factor coupling characteristics through a preset driving risk field model for the merge area; determining the vehicle interaction potential field strength generated by vehicle interaction in the highway merge area based on the steering angle, the vehicle type, the relative speed, and the relative distance.

[0087] It can be understood that the above vehicle interaction potential field strength can be the binding force or repulsive force generated during vehicle interaction. In this embodiment, the vehicle interaction potential field strength E vehicle is calculated based on data such as the steering angle of the merging vehicle, vehicle type, relative speed and relative distance from surrounding vehicles, and driver characteristics. The formula is:

[0088]

[0089] In the formula, Δd is the relative distance between the merging vehicle and surrounding vehicles in the merge area, Δν is the relative speed between the merging vehicle and surrounding vehicles in the merge area, δ is the steering angle, is the vehicle type, S d is the driving style factor, F d is the driver fatigue factor, α is the speed difference influence coefficient, β is the steering angle influence coefficient, K v is the vehicle interaction coefficient, is the vehicle type adjustment factor.

[0090] Correspondingly, the step S202 includes:

[0091] Step S202b: determining the driving risk situation value corresponding to the highway merge area based on the road boundary potential field strength, the lane line potential field strength, the construction area potential field strength, and the vehicle interaction potential field strength through the potential field superposition theory.

[0092] In this embodiment, the road boundary potential field strength, the lane line potential field strength, the construction area potential field strength, and the vehicle interaction potential field strength can be superimposed based on the potential field superposition theory, and the driving risk situation value can be calculated based on the superimposed potential field strength.

[0093] Specifically, the step S202b includes: superimposing the road boundary potential field strength, the lane line potential field strength, the construction area potential field strength, and the vehicle interaction potential field strength through the potential field superposition theory to obtain the combined potential field strength; determining the driving risk situation value corresponding to the highway merge area based on a preset reference risk value, a preset risk growth coefficient, and the combined potential field strength.

[0094] It should be understood that the above synthetic potential field intensity can be the potential field intensity obtained by superimposing the potential field intensity of the road boundary, the potential field intensity of the lane line, the potential field intensity of the construction area, and the potential field intensity of vehicle interaction.

[0095] It can be understood that the above preset reference risk value can be the basic risk level of the highway merging area under the ideal traffic state without external interference; the above preset risk growth value can be the value used to quantify the superimposed effect of different risk factors on the reference risk value.

[0096] In practical applications, the device can define a weight ω based on the potential field intensity of each element in the highway merging area on the merging vehicle. i Calculate the synthetic potential field intensity by synthesizing each potential field intensity according to the potential field superposition principle. The calculation formula can be:

[0097]

[0098] In the formula, E total is the synthetic potential field intensity, E i is the potential field intensity of each element in the highway merging area on the merging vehicle, and ω i is the weight corresponding to the potential field intensity.

[0099] After calculating the synthetic potential field intensity E total in this embodiment, the reference risk value and the risk growth coefficient can be defined, and the driving risk situation value R can be calculated based on the synthetic potential field intensity. The calculation formula can be:

[0100]

[0101] In the formula, R0 is the reference risk value, and γ is the risk growth coefficient.

[0102] The step S40 includes:

[0103] Step S40': Determine the high-risk scenarios in the highway merging area based on the scenario clustering result and the driving risk situation value.

[0104] In this embodiment, the device can calculate the average risk level of each scenario in the merging area based on the scenario clustering result and the driving risk situation value of the highway merging area, and finally introduce the maximum inter-class variance method to determine the high-risk scenarios in the highway merging area based on this average risk level.

[0105] In this embodiment, it is disclosed that the potential field intensity generated by several elements in the highway merging area is determined based on the multi-factor coupling characteristics through a preset driving risk field model in the merging area. The potential field intensity is the potential field intensity of each element on the merging vehicle in the highway merging area; the driving risk situation value corresponding to the highway merging area is determined according to the potential field intensity and the potential field superposition theory; the high-risk scenarios in the highway merging area are determined based on the scenario clustering result and the driving risk situation value; since in this embodiment, the potential field intensity generated by each element in the highway merging area can be superimposed through the potential field superposition theory, and the driving risk situation value is determined based on the synthesized potential field intensity obtained after the superposition, and finally the high-risk scenarios in the highway merging area are determined based on the scenario clustering result and the driving risk situation value of the merging area, the constraint and guiding effects of the road boundary, lane lines, construction area on the vehicle and the interaction between vehicles can be comprehensively considered, the risk situation of the merging area can be comprehensively evaluated, and thus the accuracy of identifying the high-risk state in the merging area can be improved.

[0106] Based on the first embodiment and / or the second embodiment of the present application, in the third embodiment of the present application, for the same or similar content as the above embodiments, reference may be made to the above introduction and will not be repeated hereinafter. On this basis, please refer to Figure 3 , Figure 3 which is a schematic flow chart provided for the third embodiment of the highway risk assessment method of the present application.

[0107] In this embodiment, step S30 further includes steps S301 to S303:

[0108] Step S301: Perform low-dimensional mapping on the multi-factor coupling characteristics through a preset risk scenario clustering model in the merging area to obtain low-dimensional multi-factor coupling characteristics.

[0109] It can be understood that the above low-dimensional multi-factor coupling characteristics can be the characteristics obtained after mapping the multi-factor coupling characteristics to a low-dimensional space.

[0110] Step S302: Restore the low-dimensional multi-factor coupling characteristics to the original feature space to obtain the restored multi-factor coupling characteristics.

[0111] It should be understood that the above restored multi-factor coupling characteristics can be the characteristics obtained after restoring the low-dimensional multi-factor coupling characteristics to the original feature space.

[0112] Step S303: Adopt a soft assignment method to perform scenario clustering on the restored multi-factor coupling characteristics to obtain a scenario clustering result.

[0113] In practical applications, the device can perform scenario clustering on the multi-factor coupling features through the Deep Embedded Clustering (DEC) algorithm in the preset confluence area risk scenario clustering model. Specifically, the encoder can be used to map the high-dimensional data x of the confluence area i to a low-dimensional space to obtain low-dimensional multi-factor coupling features z i . Then, the decoder is used to restore the low-dimensional multi-factor coupling features z i to the original feature space and cluster through the soft assignment method. Define the t-distribution as the probability distribution and calculate the probability q that a sample belongs to a certain cluster ij . Finally, the scenario clustering result is obtained. Among them, the deep embedded clustering algorithm is an unsupervised learning method that combines deep learning and clustering analysis. It can learn the low-dimensional embedded representation of data through a deep neural network and perform clustering in this embedded space.

[0114] In this embodiment, in order to improve the accuracy of scenario clustering, the network of the preset confluence area risk scenario clustering model can also be updated. Specifically, first, the reconstruction loss function can be defined according to the model input data and the output of the autoencoder to optimize the autoencoder to learn the key features of the data. The formula can be:

[0115]

[0116] Then, according to the soft assignment result of the model and the target distribution p ij are matched, and the loss function is defined using the KL divergence. The formula can be:

[0117]

[0118] In this embodiment, by minimizing and the key information of the driving scenario data can be retained, noise interference can be reduced, and end-to-end optimization can be performed in the low-dimensional space.

[0119] Furthermore, the step S40' includes: determining the average risk level corresponding to each clustering scenario in the highway confluence area based on the scenario clustering result and the driving risk situation value; determining the optimal dynamic risk threshold according to the average risk level; and determining the high-risk scenarios in the highway confluence area based on the optimal dynamic risk threshold.

[0120] In this embodiment, the device can calculate the average risk level corresponding to each confluence area scenario cluster based on the output result of the preset confluence area risk scenario clustering model, that is, the scenario clustering result, in combination with the driving risk situation value. The calculation formula can be:

[0121]

[0122] In the formula, is the scene cluster of each clustering scene, and S j is the scene cluster corresponding to the average risk level, and R i is the driving risk situation value.

[0123] After calculating the average risk level corresponding to each clustering scene, based on this average risk level, the Otsu method can be used to achieve adaptive threshold segmentation, and the threshold corresponding to the maximum between-class variance is selected as the optimal dynamic risk threshold τ * , and all clustering categories exceeding this threshold are screened as high-risk scenes The formula can be:

[0124]

[0125] Among them, the Otsu method is a global image threshold segmentation algorithm, which can binarize the image into foreground and background by automatically calculating the optimal segmentation threshold of the image, and is widely used in the fields of image processing, computer vision and pattern recognition.

[0126] In this embodiment, the maximum between-class variance method can be introduced to determine high-risk scenes, automatically calculate the threshold to dynamically distinguish high-risk scenes from low-risk scenes, and determine the high-risk situation threshold in the highway merging area, so as to provide active warnings for the assisted driving system, enabling it to predict risks and respond in a timely manner when facing high-risk scenes in the merging area, and at the same time providing strong support for the safe and efficient application of intelligent driving vehicles on the highway.

[0127] In this embodiment, it is disclosed that a multi-factor coupling feature is mapped to a low dimension through a preset risk scene clustering model in the highway merging area to obtain a low-dimensional multi-factor coupling feature; the low-dimensional multi-factor coupling feature is restored to the original feature space to obtain a restored multi-factor coupling feature; a soft assignment method is used to perform scene clustering on the restored multi-factor coupling feature to obtain a scene clustering result; since this embodiment can use the soft assignment method to perform scene clustering on the restored multi-factor coupling feature to obtain a scene clustering result, the scene clusters corresponding to each risk scene in the highway merging area can be accurately determined, so that the high-risk state during the merging process can be more accurately identified subsequently, thereby improving the traffic safety and stable operation of the merging area.

[0128] It should be noted that the above examples are only for understanding this application and do not constitute a limitation to the highway risk assessment method of this application. Based on this technical concept, more forms of simple transformations are within the protection scope of this application.

[0129] This application also provides a highway risk assessment device. Please refer toFigure 4 , the highway risk assessment device includes:

[0130] A data acquisition module 10, configured to acquire vehicle trajectory data and road environment data corresponding to a target section in a highway merging area;

[0131] A feature construction module 20, configured to construct multi-factor coupling features of the highway merging area based on the vehicle trajectory data and the road environment data;

[0132] A scenario clustering module 30, configured to perform scenario clustering on the multi-factor coupling features through a preset risk scenario clustering model for the merging area to obtain a scenario clustering result;

[0133] A risk assessment module 40, configured to determine high-risk scenarios in the highway merging area based on the scenario clustering result.

[0134] The highway risk assessment device provided in this application adopts the highway risk assessment method in the above embodiment, and can solve the technical problems that the risk assessment method in the prior art is difficult to accurately identify high-risk states during driving and cannot give early warnings for high-risk scenarios in a timely manner. Compared with the prior art, the beneficial effects of the highway risk assessment device provided in this application are the same as those of the highway risk assessment method provided in the above embodiment, and other technical features in the highway risk assessment device are the same as those disclosed in the method of the above embodiment, and will not be elaborated here.

[0135] This application provides a highway risk assessment device. The highway risk assessment device includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein, the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the highway risk assessment method in Embodiment 1 above.

[0136] Refer to the following Figure 5 , which shows a schematic structural diagram of a highway risk assessment device suitable for implementing the embodiments of this application. The highway risk assessment device in the embodiments of this application may include, but is not limited to, mobile terminals such as mobile phones, laptop computers, digital broadcast receivers, PDAs (Personal Digital Assistants), PADs (Portable Application Descriptions), PMPs (Portable Media Players), vehicle-mounted terminals (such as vehicle-mounted navigation terminals), etc., and fixed terminals such as digital TVs, desktop computers, etc. Figure 5The shown highway risk assessment device is merely an example and should not impose any limitations on the functions and usage scope of the embodiments of this application.

[0137] As Figure 5 shown, the highway risk assessment device may include a processing device 1001 (such as a central processing unit, a graphics processing unit, etc.), which can perform various appropriate actions and processes according to the program stored in the read-only memory 1002 or the program loaded from the storage device 1003 into the random access memory 1004. In the random access memory 1004, various programs and data required for the operation of the highway risk assessment device are also stored. The processing device 1001, the read-only memory 1002, and the random access memory 1004 are connected to each other through a bus 1005. The input / output interface 1006 is also connected to the bus. Generally, the following systems can be connected to the input / output interface 1006: an input device 1007 including, for example, a touch screen, a touchpad, a keyboard, a mouse, an image sensor, a microphone, an accelerometer, a gyroscope, etc.; an output device 1008 including, for example, a liquid crystal display (LCD: Liquid Crystal Display), a speaker, a vibrator, etc.; a storage device 1003 including, for example, a magnetic tape, a hard disk, etc.; and a communication device 1009. The communication device 1009 can allow the highway risk assessment device to communicate with other devices wirelessly or wiredly to exchange data. Although the figure shows a highway risk assessment device with various systems, it should be understood that it is not required to implement or have all the shown systems. More or fewer systems can be alternatively implemented or had.

[0138] Specifically, according to the embodiments disclosed in this application, the process described above with reference to the flowchart can be implemented as a computer software program. For example, the embodiments disclosed in this application include a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program contains program codes for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from the network through the communication device, or installed from the storage device 1003, or installed from the read-only memory 1002. When the computer program is executed by the processing device 1001, the above-mentioned functions defined in the method of the embodiments disclosed in this application are executed.

[0139] The highway risk assessment device provided by this application adopts the highway risk assessment method in the above-mentioned embodiment and can solve the technical problem of highway risk assessment. Compared with the prior art, the beneficial effects of the highway risk assessment device provided by this application are the same as those of the highway risk assessment method provided by the above-mentioned embodiment, and other technical features in this highway risk assessment device are the same as those disclosed in the method of the previous embodiment, which will not be elaborated here.

[0140] It should be understood that each part disclosed in this application can be implemented by hardware, software, firmware, or a combination thereof. In the description of the above embodiments, specific features, structures, materials, or characteristics can be combined in a suitable manner in any one or more embodiments or examples.

[0141] As mentioned above, the above are only specific embodiments of this application, but the protection scope of this application is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed in this application, and all should be covered by the protection scope of this application. Therefore, the protection scope of this application should be subject to the protection scope of the claims.

[0142] This application provides a computer-readable storage medium having computer-readable program instructions (i.e., computer programs) stored thereon, and the computer-readable program instructions are used to execute the highway risk assessment method in the above embodiments.

[0143] The computer-readable storage medium provided by this application can be, for example, a USB flash drive, but is not limited to electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems or devices, or any combination of the above. More specific examples of computer-readable storage media can include, but are not limited to: electrical connections with one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM) or flash memory, optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the above. In this embodiment, the computer-readable storage medium can be any tangible medium that contains or stores a program, and the program can be used by or in conjunction with an instruction execution system or device. The program code contained on the computer-readable storage medium can be transmitted by any appropriate medium, including but not limited to: wires, optical cables, RF (Radio Frequency), etc., or any suitable combination of the above.

[0144] The above computer-readable storage medium can be included in the highway risk assessment device; or it can exist alone without being assembled into the highway risk assessment device.

[0145] The above computer-readable storage medium carries one or more programs, which, when executed by the highway risk assessment device, cause the highway risk assessment device to: collect vehicle trajectory data and road environment data corresponding to a target section in a highway merging area; construct a multi-factor coupling feature of the highway merging area based on the vehicle trajectory data and the road environment data; perform scenario clustering on the multi-factor coupling feature through a preset merging area risk scenario clustering model to obtain a scenario clustering result; and determine a high-risk scenario in the highway merging area based on the scenario clustering result.

[0146] Computer program code for performing the operations of the present application may be written in one or more programming languages or combinations thereof. The programming languages include object-oriented programming languages such as Java, Smalltalk, C++, and also include conventional procedural programming languages such as the "C" language or similar programming languages. The program code may execute entirely on the user's computer, partially on the user's computer, execute as a stand-alone software package, execute partially on the user's computer and partially on a remote computer, or execute entirely on the remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., by connecting through an Internet service provider using the Internet).

[0147] The flowcharts and block diagrams in the accompanying drawings illustrate the possible architectures, functions, and operations of systems, methods, and computer program products according to various embodiments of the present application. In this regard, each block in the flowchart or block diagram may represent a module, a program segment, or a part of code that contains one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions marked in the blocks may occur in a different order than marked in the accompanying drawings. For example, two consecutive blocks shown may actually be executed substantially in parallel, and they may sometimes be executed in the reverse order, depending on the functions involved. It should also be noted that each block in the block diagram and / or flowchart, and the combinations of blocks in the block diagram and / or flowchart, may be implemented by a dedicated hardware-based system for performing the specified functions or operations, or may be implemented by a combination of dedicated hardware and computer instructions.

[0148] The modules described in the embodiments of the present application may be implemented in software or in hardware. Wherein, the name of the module does not constitute a limitation to the unit itself in some cases.

[0149] The readable storage medium provided by this application is a computer-readable storage medium, and the computer-readable storage medium stores computer-readable program instructions (i.e., computer programs) for executing the above-mentioned highway risk assessment method, which can solve the technical problems that the risk assessment method in the prior art is difficult to accurately identify high-risk states during driving and cannot give early warnings for high-risk scenarios in a timely manner. Compared with the prior art, the beneficial effects of the computer-readable storage medium provided by this application are the same as those of the highway risk assessment method provided in the above embodiments, and will not be elaborated here.

[0150] The above are only partial embodiments of this application, and do not limit the patent scope of this application. Any equivalent structural transformation made under the technical concept of this application by using the content of the specification and drawings of this application, or direct / indirect application in other related technical fields, is included in the patent protection scope of this application.

Claims

1. A highway risk assessment method, characterized in that: The method includes: Collect vehicle trajectory data and road environment data corresponding to the target section in the merging area of ​​the highway; Constructing a multi-factor coupling feature of the highway merging area based on the vehicle trajectory data and the road environment data; The multi-factor coupling characteristics are clustered by using a preset merging area risk scenario clustering model to obtain a scenario clustering result; A high-risk scene in the highway merging area is determined based on the scene clustering result.

2. The method according to claim 1, characterized in that Before the step of clustering the multi-factor coupling features by using a preset merging area risk scenario clustering model to obtain a scenario clustering result, the step further includes: Determine the potential field strength generated by several elements in the merging area of ​​the expressway based on the multi-element coupling characteristics by presetting the merging area driving risk field model, wherein the potential field strength is the potential field strength of each element to vehicles merging into the merging area of ​​the expressway; Determine the driving risk situation value corresponding to the highway merging area according to the potential field strength and potential field superposition theory; The step of determining the high-risk scene in the highway merging area based on the scene clustering result includes: A high-risk scene in the highway merging area is determined based on the scene clustering result and the driving risk situation value.

3. The method according to claim 2, characterized in that The potential field strength includes: road boundary potential field strength, lane line potential field strength and construction area potential field strength; the step of determining the potential field strength generated by several elements in the merging area of ​​the highway based on the multi-element coupling characteristics by presetting the merging area driving risk field model includes: By presetting the merging area driving risk field model, the lane line distance between the vehicle traveling in the merging area of ​​the expressway and the lane center line, the current speed of the vehicle, the road speed threshold, and the boundary distance, safety distance and direction angle between the vehicle and the construction area boundary are determined according to the multi-factor coupling characteristics; Determine a road boundary potential field strength corresponding to a road boundary in the merging area of ​​the expressway and a lane line potential field strength corresponding to a lane line based on the lane line distance, the current speed of the vehicle and the road travel speed threshold; Determine the potential field strength of the construction area corresponding to the construction area in the merging area of ​​the expressway based on the boundary distance, the safety distance and the direction angle; The step of determining the driving risk situation value corresponding to the highway merging area according to the potential field strength and potential field superposition theory includes: The driving risk situation value corresponding to the merging area of ​​the expressway is determined based on the road boundary potential field strength, the lane line potential field strength and the construction area potential field strength through potential field superposition theory.

4. The method according to claim 3, characterized in that The potential field strength also includes: vehicle interaction potential field strength; the step of determining the potential field strength generated by several elements in the merging area of ​​the highway based on the multi-element coupling characteristics by presetting the merging area driving risk field model includes: Determine the steering angle, vehicle type, and relative speed and relative distance of the vehicle merging into the merging area of ​​the highway to the surrounding vehicles based on the multi-factor coupling characteristics by presetting the merging area driving risk field model; determining a vehicle interaction potential field strength generated by vehicle interaction in the highway merging area based on the steering angle, the vehicle type, the relative speed, and the relative distance; The step of determining the driving risk situation value corresponding to the highway merging area based on the road boundary potential field strength, the lane line potential field strength and the construction area potential field strength through potential field superposition theory includes: The driving risk situation value corresponding to the highway merging area is determined based on the road boundary potential field strength, the lane line potential field strength, the construction area potential field strength and the vehicle interaction potential field strength through potential field superposition theory.

5. The method according to claim 4, characterized in that The step of determining the driving risk situation value corresponding to the highway merging area based on the road boundary potential field strength, the lane line potential field strength, the construction area potential field strength and the vehicle interaction potential field strength through potential field superposition theory includes: The road boundary potential field strength, the lane line potential field strength, the construction area potential field strength and the vehicle interaction potential field strength are superimposed by potential field superposition theory to obtain a composite potential field strength; The driving risk situation value corresponding to the highway merging area is determined based on a preset baseline risk value, a preset risk growth coefficient and the synthetic potential field strength.

6. The method according to claim 1, characterized in that The step of performing scene clustering on the multi-factor coupling features by using a preset merging area risk scene clustering model to obtain scene clustering results includes: By presetting a confluence area risk scenario clustering model, the multi-factor coupling characteristics are low-dimensionally mapped to obtain low-dimensional multi-factor coupling characteristics; Restoring the low-dimensional multi-factor coupling feature to the original feature space to obtain a restored multi-factor coupling feature; A soft allocation method is used to perform scene clustering on the restored multi-factor coupling features to obtain a scene clustering result.

7. The method according to claim 2, characterized in that The step of determining the high-risk scene in the highway merging area based on the scene clustering result and the driving risk situation value comprises: Determining an average risk level corresponding to each clustered scene in the merging area of ​​the highway based on the scene clustering result and the driving risk situation value; determining an optimal dynamic risk threshold according to the average risk level; A high-risk scenario in the highway merging area is determined based on the optimal dynamic risk threshold.

8. A highway risk assessment device, characterized in that: The device comprises: A data collection module is used to collect vehicle trajectory data and road environment data corresponding to the target section in the merging area of ​​the expressway; A feature construction module, used to construct a multi-factor coupling feature of the highway merging area based on the vehicle trajectory data and the road environment data; A scenario clustering module is used to perform scenario clustering on the multi-factor coupling features by using a preset merging area risk scenario clustering model to obtain a scenario clustering result; A risk assessment module is used to determine high-risk scenes in the highway merging area based on the scene clustering results.

9. A highway risk assessment device, characterized in that: The device comprises: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the computer program is configured to implement the steps of the highway risk assessment method according to any one of claims 1 to 7.

10. A storage medium, characterized in that: The storage medium is a computer-readable storage medium, and a computer program is stored on the storage medium. When the computer program is executed by a processor, the steps of the highway risk assessment method according to any one of claims 1 to 7 are implemented.

Images