Traffic risk discrimination method and equipment for highway construction operation section
By combining macroscopic and microscopic analysis methods, utilizing network distance attenuation coefficients and UAV aerial photography technology, and integrating ground and air source information, the problem of difficulty in identifying traffic risks on highway construction sites has been solved, enabling more comprehensive risk assessment and management.
Patent Information
- Application Number
- CN202511355106.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-09-22
- Publication Date
- 2026-01-13
AI Technical Summary
Existing methods for assessing traffic risks on highway construction sites are insufficient to comprehensively consider both macro and micro factors, resulting in fragile traffic safety and making it difficult to form a comprehensive risk assessment.
A macro-level analysis is conducted by combining network distance attenuation coefficient, shortest path betweenness sharing rate, and neighborhood traffic fluctuation transmission coefficient, while a micro-level analysis is conducted by combining drone aerial video. Finally, a comprehensive assessment of traffic risks is made by integrating ground source and air source information.
It enables a comprehensive characterization of traffic risks on highway construction sites, provides real-time data support and risk control measures, and improves the reliability and practicality of risk assessment.
Smart Images
Figure CN121330901A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of traffic risk analysis, and in particular to a method and equipment for identifying traffic risks on highway construction sections. Background Technology
[0002] Highway construction work occupies some lanes on the affected section of the road, impacting traffic safety risks. From a macro perspective, when the overall capacity of the construction area decreases, its ability to withstand fluctuations in traffic flow decreases, thus increasing the likelihood of increased traffic risks. From a micro perspective, traffic risks are influenced by driving behavior. Vehicles inevitably need to overtake and change lanes when passing through construction areas, leading to increased dangerous driving behaviors and a higher probability of traffic conflicts in the area.
[0003] In macro-level road risk sensitivity, the accident rate is significantly positively correlated with traffic volume. When maintenance work occupies part of the lanes, the road section's capacity immediately decreases, creating a "bottleneck effect." If the surrounding diversion roads are saturated at this time, vehicles will accumulate upstream of the work area, and the increased density will cause the accident risk to increase exponentially.
[0004] Among the micro-level risks of abnormal driving behavior, increased traffic flow leads to a significant deterioration in driving behavior around the work area, mainly manifested in three aspects: first, reduced headway; second, a surge in lane-changing frequency, especially in the transition section of the work area, where the average number of lane changes per vehicle is several times that of normal road sections; and third, increased speed dispersion. These factors also lead to an increase in traffic risks in the construction work area.
[0005] Therefore, both macro and micro factors will make traffic safety in construction work areas more vulnerable. Summary of the Invention
[0006] This application provides a method and equipment for traffic risk assessment of highway construction sections, which addresses the following technical problem: In existing traffic risk assessments of highway construction sections, both macroscopic and microscopic factors make traffic safety in the construction area more vulnerable, making it difficult to form a comprehensive traffic risk assessment result for the construction section under traffic safety management.
[0007] The embodiments of this application adopt the following technical solutions: On one hand, this application provides a method for traffic risk assessment of highway construction sections, including: calculating the correlation between the construction section and the remaining supplementary sections in the highway network based on the network distance attenuation coefficient and the shortest path betweenness sharing rate in the traffic network to obtain the correlation neighborhood of the construction section; calculating the fluctuation transmission of traffic flow in the neighboring sections of the correlation neighborhood to obtain the neighborhood traffic flow fluctuation transmission coefficient; and normalizing the enhanced risk sensitivity coefficient of the construction section under the relevant lane capacity based on the neighborhood traffic flow fluctuation transmission coefficient to obtain the road in the construction section. Risk sensitivity; using drone aerial photography, based on traffic parameters corresponding to the non-construction and construction phases, the deviation of traffic parameters for the same road segment under construction and non-construction conditions is calculated, and the deviation of abnormal driving behavior is obtained; the deviation of abnormal driving behavior and the coefficient of the impact range of abnormal driving behavior on traffic accidents caused by abnormal driving behavior are used to calculate the behavioral risk contribution index, and the abnormal driving behavior risk contribution index is determined; the road risk sensitivity based on ground source information and the abnormal driving behavior risk contribution index based on air source information are fused to obtain the comprehensive traffic risk judgment result of the construction section.
[0008] This application's embodiments utilize traffic flow data obtained from ground-based ETC detectors to conduct macro-level traffic risk analysis. This macro-level analysis reflects the sensitivity of traffic risk in construction sections with reduced capacity to fluctuations in traffic flow from neighboring sections. Simultaneously, micro-level traffic risk analysis is conducted using aerial video footage from drones over the construction area. This micro-level analysis reflects the adverse impact of abnormal driving behavior caused by road construction on traffic risk. Finally, the macro- and micro-level analysis results are integrated to obtain a comprehensive assessment of the traffic risk of the construction section. The method described in this patent provides a more comprehensive characterization of traffic risks in highway construction sections, offering important theoretical guidance and data support for optimizing highway operation and management and ensuring the safety of construction workers.
[0009] In one feasible implementation, before calculating the correlation between construction sections and remaining complement sections in the highway network based on the network distance attenuation coefficient and shortest path betweenness sharing rate in the traffic network to obtain the correlation neighborhood of the construction sections, the method further includes: determining the highway network based on the connections between network nodes; wherein the network nodes include at least: toll stations and interchanges; dividing the set of road segments in the highway network into a set of construction sections and a set of non-construction sections; according to The network distance attenuation coefficient is obtained. ;in, Indicates the construction work section and the remaining supplementary road segments Topological distance in a road network; It is a mathematical constant; according to The shortest path betweenness sharing rate is obtained. ;in, This indicates that the highway network includes the road sections where construction work is being carried out. The number of shortest paths; This indicates that the remaining supplementary road segments are included in the highway network. The number of shortest paths; This indicates that the highway network also includes the aforementioned construction sections. and the remaining supplementary road segments The number of shortest paths; the shortest path betweenness sharing rate reflects the number of shortest paths in the construction section. and the remaining supplementary road segments The co-occurrence frequency in the shortest path of the road network, reflecting the strength of the indirect correlation between the two.
[0010] In one feasible implementation, based on the network distance attenuation coefficient and the shortest path betweenness sharing rate in the traffic network, the correlation between the construction operation segment and the remaining supplementary segment in the highway network is calculated to obtain the correlation neighborhood of the construction operation segment. Specifically, this includes: based on... The construction section was obtained. With the remaining supplementary road segments exist The passing of time Traffic time-series correlation within hours ;in, Indicates the construction work section exist Flow rate at any moment ; Indicates the construction work section in the past The average flow rate per hour; Indicates the remaining supplementary road segment exist Flow rate at any moment ; Indicates the remaining supplementary collection section in the past The average flow rate per hour; =24 ;according to The construction section was obtained. With the remaining supplementary road segments exist Time-related correlation ;in, , , ; This is the network distance attenuation coefficient; The shortest path betweenness sharing rate; according to ,get Set of correlation indicators at different times ;in, The construction section is the section of road where the work is being carried out. and the remaining supplementary road segments exist Correlation at time; for the set of correlation indicators The elements are sorted from largest to smallest to obtain the construction section. exist ordered sequence of time ;according to , get in Correlation neighborhood at time step ;in, K The ordered sequence The front of the middle K The largest element; x For the remaining supplementary road segments Any section of the road.
[0011] In one feasible implementation, traffic flow fluctuation transmission calculations are performed on neighboring road segments within the relevant neighborhood to obtain a neighborhood traffic flow fluctuation transmission coefficient. Specifically, this includes: based on... ,get Neighborhood flow fluctuation propagation coefficient at time 1 ;in, K For ordered sequences The front of the middle K The largest element; In the relevance neighborhood, the first The adjacent road section The flow of time, and ; In the relevance neighborhood, the first The adjacent road section in the past The average flow rate per hour.
[0012] In one feasible implementation, based on the neighborhood traffic fluctuation transmission coefficient, the enhanced risk sensitivity coefficient of the construction section under the relevant lane capacity is normalized to obtain the road risk sensitivity of the construction section, specifically including: according to The construction section was obtained. The sensitivity coefficient of the enhanced sensitivity to risks at all times ;in, Indicates the construction work section The original traffic capacity before construction. The construction section is the section of road where the work is being carried out. Road traffic capacity during construction; The construction section exist The neighborhood flow fluctuation propagation coefficient at time t; The construction section The basic risk sensitivity coefficient, and the basic risk sensitivity coefficient This reflects road sections under construction even when they are not currently under construction. The sensitivity of traffic risk between historical traffic flow data and historical accident rate data per unit of vehicle flow; based on The construction section was obtained. exist The road risk sensitivity at any given time ;in, ,and This represents the maximum value of the set of sensitivity coefficients for enhanced sensitive risks corresponding to all construction operation sections.
[0013] In one feasible implementation, before calculating the deviation of traffic parameters for the same road segment under construction and non-construction conditions based on traffic parameters corresponding to the non-construction and construction phases using aerial photography from a drone, and before determining the deviation of abnormal driving behavior, the method further includes: using a YOLO+DeepSort deep learning model and a LaneNet deep learning model to identify and locate vehicle movement trajectories and highway lane lines in the aerial video captured by the drone from above; according to The lane keeping deviation was obtained. ;in, For highway lane width, The distance between the vehicle's centerline and the centerline of its lane is defined as follows: Non-construction traffic parameters are obtained and calculated for non-construction sections of road during the non-construction phase; these non-construction traffic parameters include: non-construction average speed, non-construction average acceleration, non-construction average headway, and non-construction average lane keeping deviation; Construction traffic parameters are obtained and calculated for construction sections of road during the construction phase; these construction traffic parameters include: construction average speed, construction average acceleration, construction average headway, and construction average lane keeping deviation; wherein, the non-construction phase and the construction phase refer to different stages of the same road segment.
[0014] In one feasible implementation, by using aerial photography from a drone, based on traffic parameters corresponding to the non-construction and construction phases, the deviation of traffic parameters for the same road segment under construction and non-construction conditions is calculated, and the deviation of abnormal driving behavior is derived, specifically including: according to ,get Speed deviation at time ;in, for The average construction speed under the specified construction conditions at any given time; The non-construction average speed under the non-construction state; according to ,get acceleration deviation at time ;in, for The average acceleration during construction under the specified construction conditions at any given time; The non-construction average acceleration in the non-construction state; according to ,get Camber deviation at any given moment ;in, for The average distance between construction vehicles under the specified construction conditions at any given time; The non-construction vehicle head spacing in the non-construction state; according to ,get Lane keeping deviation at any time ;in, for The average lane keeping deviation during construction under the specified construction conditions at any given time; The non-construction average lane keeping deviation under the non-construction state; according to ,get Deviation of the abnormal driving behavior at any given time ;in, , , , .
[0015] In one feasible implementation, the deviation degree of the abnormal driving behavior and the coefficient of the impact range of the abnormal driving behavior in traffic accidents caused by the abnormal driving behavior are used to calculate the behavioral risk contribution index, thereby determining the abnormal driving behavior risk contribution index. Specifically, this includes: based on... ,get The influence range coefficient index of the abnormal driving behavior at the given time. ;in, This represents the average vehicle density of traffic flow on a road section under non-construction conditions over a historical period. This indicates the average vehicle density of traffic flow on the road section under construction over a historical period. according to ,get Risk contribution index of abnormal driving behavior at the specified time. ;in, for Deviation of the aforementioned abnormal driving behavior at any given time; , .
[0016] In one feasible implementation, the road risk sensitivity based on ground-source information and the abnormal driving behavior risk contribution index based on air-source information are fused to obtain a comprehensive traffic risk assessment result for the construction section, specifically including: based on ,get The comprehensive traffic risk assessment results at the current time ;in, In order to be in The road risk sensitivity obtained based on geospatial information at any given time; In order to be in The risk contribution index of the abnormal driving behavior obtained based on the spatial source information at any given time; , .
[0017] On the other hand, this application also provides a traffic risk assessment device for highway construction sections, including: 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, so that the at least one processor can execute a traffic risk assessment method for highway construction sections as described in any of the above embodiments.
[0018] This application provides a method and equipment for traffic risk assessment in highway construction sections. Compared with the prior art, the embodiments of this application have the following beneficial technical effects: 1. By combining factors such as network distance attenuation coefficient, shortest path betweenness sharing rate, and neighborhood traffic fluctuation transmission coefficient, the traffic risk of construction work sections can be assessed more accurately.
[0019] 2. By calculating the transmission of traffic flow fluctuations in neighboring road sections, changes in traffic flow near construction sites can be monitored in real time, providing timely data support for traffic management.
[0020] 3. By normalizing the enhanced risk sensitivity coefficient under lane capacity, lane capacity can be managed more effectively, reducing traffic congestion and accident risks caused by construction.
[0021] 4. By calculating road risk sensitivity, it is possible to identify which road sections and conditions are more susceptible to abnormal driving behavior, thereby enabling the implementation of corresponding risk control measures.
[0022] 5. By utilizing drone aerial photography technology, traffic parameter data can be collected efficiently and comprehensively, improving the accuracy and efficiency of data collection.
[0023] 6. By comparing traffic parameters during the construction phase and before the construction phase, the impact of construction on traffic flow can be more clearly identified, providing a basis for optimizing the construction plan.
[0024] 7. By calculating the deviation of traffic parameters and the deviation of abnormal driving behavior, abnormal driving behavior can be identified, which helps to prevent and reduce traffic accidents.
[0025] 8. By calculating the risk contribution index of abnormal driving behavior, a quantitative basis can be provided for the intervention and punishment of abnormal driving behavior.
[0026] 9. Integrating the road risk sensitivity of ground source information with the abnormal driving behavior risk contribution index of air source information can provide more comprehensive risk assessment results and improve the reliability and practicality of risk identification. Attached Figure Description
[0027] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. In the drawings: Figure 1 A flowchart illustrating a traffic risk assessment method for highway construction sections provided in this application embodiment; Figure 2 This is a schematic diagram of the structure of a traffic risk assessment device for a highway construction section, provided as an embodiment of this application. Detailed Implementation
[0028] To enable those skilled in the art to better understand the technical solutions in this application, the technical solutions in the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this specification, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of this application.
[0029] This application provides a method for traffic risk assessment in highway construction sections, such as... Figure 1 As shown, the method for assessing traffic risks on highway construction sections specifically includes steps S101-S106: S101. Based on the network distance attenuation coefficient and the shortest path betweenness sharing rate in the traffic network, the correlation between the construction operation section and the remaining supplementary section in the expressway network is calculated to obtain the correlation neighborhood of the construction operation section.
[0030] It should be noted that this application will model the impact of fluctuations in traffic flow in the vicinity of the construction section and the reduction in the capacity of the construction section itself on traffic risk. First, the relevant neighborhood of the construction section is determined, and the transmission coefficient of neighborhood traffic flow fluctuation is determined accordingly. Then, the basic risk sensitivity coefficient and the enhanced risk sensitivity coefficient are calculated. Finally, the normalized enhanced risk sensitivity coefficient is used as the Road Risk Sensitivity (RTS) index to measure the impact of fluctuations in traffic flow in the vicinity of the construction section and the reduction in the capacity of the construction section itself on traffic risk.
[0031] Specifically, the highway network needs to be determined first based on the connections between the network nodes. These network nodes include, at a minimum, toll stations and interchanges.
[0032] Furthermore, the set of road segments in the expressway network is divided into a set of road segments under construction and a set of road segments not under construction.
[0033] In one embodiment, a highway network is a network composed of road network nodes (including toll stations, interchanges, etc.) and the connections between nodes (i.e., road segments). Assume the highway network contains P road segments, of which N road segments are under construction and M road segments are not under construction. The set of all road segments is denoted as […]. , A subset consisting of the first N elements of a set This represents N road sections undergoing construction. A subset consisting of the last M elements of a set This indicates that road segment M is currently not under construction. It meets the following requirements. P=N+M .in, Subscripts indicate the construction stage. Subscripts indicate that the construction phase is not underway.
[0034] As a feasible implementation method, topological correlation and traffic change correlation are used to determine the "correlation neighborhood" of the construction section in the road network. For a specific construction section... The complement of the remaining P-1 roads in the entire highway network is denoted as . This application requires selecting K roads from the remaining P-1 roads to construct the construction work section based on relevance indicators. The "relevance neighborhood".
[0035] As a possible implementation method, for complement A certain section of road This refers to the remaining replenishment section. First, calculate its relationship with the construction section. The topology-related metrics (including Network Distance Attenuation Coefficient NDAC and Shortest Path Betweenness Sharing Rate SPSR) and traffic-dependent metrics (Traffic Temporal Correlation Traceability FTC) are then calculated. Construction work section correlation Then, regarding the remaining complement... Every section of road Repeat the above steps to obtain the correlation of P-1 roads, forming a set. Finally, select the set. The top K maximum values are used to construct the construction work sections, and the corresponding K road segments are selected. correlation neighborhood .
[0036] Furthermore, according to The network distance attenuation coefficient is obtained. .in, Indicates the road section under construction and remaining replenishment sections Topological distance in a road network; It is a mathematical constant.
[0037] Furthermore, according to The shortest path betweenness sharing rate is obtained. .in, This indicates that the expressway network includes sections under construction. The number of shortest paths; This indicates that the highway network includes remaining supplementary road segments. The number of shortest paths; This indicates that the highway network also includes sections under construction. and remaining replenishment sections The number of shortest paths; the shortest path betweenness sharing rate reflects the construction operation section. and remaining replenishment sections The co-occurrence frequency in the shortest path of the road network, reflecting the strength of the indirect correlation between the two.
[0038] Furthermore, according to The construction section was obtained. With remaining replenishment sections exist The passing of time Traffic time-series correlation within hours .in, Indicates the road section under construction exist Flow rate at any moment ; Indicates the road section under construction in the past The average flow rate per hour; Indicates the remaining supplementary collection section exist Flow rate at any moment ; Indicates the remaining supplementary collection section in the past The average flow rate per hour; =24 .
[0039] Furthermore, according to The construction section was obtained. With remaining replenishment sections exist Time-related correlation .in, , , ; This is the network distance attenuation coefficient; This represents the shortest path betweenness sharing rate.
[0040] Furthermore, according to ,get Set of correlation indicators at different times .in, It is a construction site. and remaining replenishment sections exist Relevance of time.
[0041] Furthermore, regarding the set of correlation indicators Sort the elements from largest to smallest to obtain the construction work sections. exist ordered sequence of time .
[0042] Furthermore, according to , get in Correlation neighborhood at time step .in, K For ordered sequences The front of the middle K The largest element;x For the remaining collection sections Any section of the road.
[0043] In one embodiment, after obtaining the construction work section correlation neighborhood In the middle, it is necessary to target the road segment set. ,get Set of correlation indicators at different times ,in It is a construction site. and remaining replenishment sections exist The correlation of time. Then for the set. Sort the elements in descending order to obtain an ordered sequence. Select the K largest elements and use their corresponding road segments as the construction work sections. exist The relevance neighborhood at time step.
[0044] S102. Perform traffic flow fluctuation transmission calculations on the neighboring road segments in the relevant neighborhood to obtain the neighborhood traffic flow fluctuation transmission coefficient.
[0045] Specifically, according to ,get Neighborhood flow fluctuation propagation coefficient at time 1 .in, K For ordered sequences The front of the middle K The largest element; Indicates the th in the relevance neighborhood The adjacent road section The flow of time, and ; Indicates the th in the relevance neighborhood The adjacent road section in the past The average flow rate per hour.
[0046] S103. Based on the neighborhood traffic fluctuation transmission coefficient, the enhanced risk sensitivity coefficient of the construction section under the relevant lane capacity is normalized to obtain the road risk sensitivity of the construction section.
[0047] It should be noted that this is a collection of highway sections. According to the road section Traffic flow under non-construction conditions Historical data and accident rate per unit of traffic volume Historical data (i.e., the ratio of accidents to traffic flow per unit time) is fitted with the following regression model: .in, , and It is a constant. called a road section The basic risk sensitivity coefficient reflects the road segment The sensitivity of traffic risk to changes in its traffic flow. When The larger the value, the more easily traffic risks are affected by changes in traffic flow.
[0048] At the same time, because RTS This reflects the sensitivity of traffic risks at the construction site to fluctuations in traffic flow in neighboring road sections from a macro perspective, exhibiting a certain degree of temporal stability. This application... Once every hour RTS The following description refers to a single update. RTS Calculation. In this application Assuming the current time is , express Sensitivity to road risks at all times.
[0049] Specifically, for the road sections under construction Its basic risk sensitivity coefficient is Lane occupancy reduces traffic capacity, leading to a more pronounced sensitivity to traffic risks. Additionally, construction work is being carried out on certain road sections. correlation neighborhood Traffic fluctuations on other roads also affect the construction site. The degree of sensitivity has a transmission effect. Therefore, the degree of sensitivity under construction conditions is called the enhanced sensitivity risk sensitivity coefficient. .
[0050] Furthermore, according to The construction site was located in The sensitivity coefficient of the enhanced sensitivity to risks at all times .in, Indicates the road section under construction The original traffic capacity before construction. It is a construction site. Road traffic capacity during construction; Road section under construction exist The neighborhood flow fluctuation propagation coefficient at time t; Road section under construction The basic risk sensitivity coefficient, and the basic risk sensitivity coefficient This reflects road sections under construction even when they are not currently under construction. The sensitivity of traffic risk between historical traffic flow data and historical accident rate data per unit of traffic flow.
[0051] As a feasible implementation method, and The calculation can be referenced in the "Technical Standards for Highway Engineering", for example: .in, This indicates the design capacity of a single lane, which is taken as 2000 vehicles / hour according to the "Technical Standards for Highway Engineering". Indicates the number of lanes in one direction. Reduction factor: .
[0052] Furthermore, according to The construction section was obtained. exist Sensitivity to road risks at all times .in, ,and This represents the maximum value of the set of sensitivity coefficients for enhanced sensitive risks across all construction sections. Furthermore, it is also necessary to ensure... Within the range of 0-1.
[0053] S104. Using drone aerial photography, based on the traffic parameters corresponding to the non-construction and construction phases, calculate the deviation of traffic parameters for the same road segment under construction and non-construction conditions, and derive the deviation of abnormal driving behavior.
[0054] It should be noted that this application will utilize drone aerial video of the construction work area to first extract basic information such as the position of moving targets and lane lines, then obtain parameters such as vehicle speed acceleration, vehicle headway, and lane keeping, and calculate the deviation degree of abnormal driving behavior (ADB_DD) and the risk contribution index of abnormal driving behavior (ADB_RCI), and finally obtain the risk contribution index of abnormal driving behavior (ADB_RCI) to model the impact of abnormal driving behavior on traffic risks of the construction work section.
[0055] Specifically, the YOLO+DeepSort deep learning model and the LaneNet deep learning model are used to identify and locate vehicle movement trajectories and highway lane lines in aerial videos captured by drones from above.
[0056] In one embodiment, this application uses a drone to take aerial photos of the transition zone and upstream transition area of the construction site, obtaining aerial video. During the filming process, the drone remains stationary with the camera pointing vertically downwards. Furthermore, this application uses the existing YOLO+DeepSort deep learning model to identify, locate, and track vehicle targets, thereby obtaining vehicle trajectories; and uses the LaneNet deep learning model to locate highway lane lines.
[0057] Furthermore, according to The lane keeping deviation was obtained. .in, For highway lane width, This is the distance by which the vehicle's centerline deviates from the centerline of its lane.
[0058] Furthermore, non-construction traffic parameters of traffic flow on non-construction operation road sections during the non-construction phase are obtained and calculated. These non-construction traffic parameters include: non-construction average speed, non-construction average acceleration, non-construction average headway, and non-construction average lane keeping deviation.
[0059] In one embodiment, a drone is used to collect aerial video during non-construction phases, extracting vehicle target and lane line information in real time, and then... Calculate the average speed during non-construction periods, using minutes as the duration (historical time period). Average acceleration Average headway and average lane keeping deviation .in, yes The average speed of all vehicles over a period of minutes; yes The average acceleration of all vehicles over a period of minutes; yes The average distance between the front and rear of all adjacent vehicles within a minute. yes All vehicles within minutes The mean.
[0060] Furthermore, based on Time's up At any given time, construction traffic parameters for the road section under construction are acquired and calculated. These parameters include: average construction speed, average construction acceleration, average headway, and average lane keeping deviation.
[0061] In one embodiment, based on Time's up Aerial video footage taken at all times is used to extract vehicle targets and lane line information, and the average speed during construction work is calculated. Average acceleration Average headway and average lane keeping deviation The non-construction phase and the construction phase refer to different stages of the same road segment, that is, the construction and non-construction operation states of the same traffic segment at different times.
[0062] Furthermore, according to ,get Speed deviation at time .in, for Average construction speed at any given moment during construction operations; This represents the average speed under non-construction conditions over a historical period. In other words, it's necessary to calculate the deviation of a series of traffic parameters between construction and non-construction conditions for the same road segment.
[0063] Furthermore, according to ,get acceleration deviation at time .in, for Average acceleration during construction operations at any given time; It represents the non-construction average acceleration during the historical period when no construction work is being carried out.
[0064] Furthermore, according to ,get Camber deviation at any given moment .in, for Average distance between construction vehicles at any given time during construction operations; This refers to the distance between non-construction vehicle heads during a historical period when they are not in a construction operation state.
[0065] Furthermore, according to ,get Lane keeping deviation at any time .in, for Average lane maintenance deviation during constant construction operations; The average lane holding deviation during the non-construction period under non-construction conditions.
[0066] Furthermore, according to ,get Deviation of abnormal driving behavior at any given time .in, , , , .
[0067] S105. Calculate the behavioral risk contribution index by combining the deviation degree of abnormal driving behavior with the coefficient of the impact range of abnormal driving behavior on traffic accidents caused by abnormal driving behavior, and determine the risk contribution index of abnormal driving behavior.
[0068] It should be noted that, due to ADB_RCIThis model reflects the impact of abnormal driving behavior on traffic risks in construction sections at a micro level, enabling more accurate modeling. Once every minute ADB_RCI The following description refers to a single update. ADB_RCI Calculation. In this application Assume the current time is , express Risk contribution index of abnormal driving behavior at any time.
[0069] Specifically, when abnormal driving behavior causes a traffic accident, the severity of the accident is related to vehicle density; the higher the vehicle density, the greater the adverse impact of the traffic accident. Therefore, this application will establish an impact range coefficient index for abnormal driving behavior based on vehicle density.
[0070] Furthermore, according to ,get The coefficient of influence range of abnormal driving behavior at any given time .in, The average vehicle density of traffic flow on road sections under non-construction conditions over a historical period; This indicates the average vehicle density of traffic flow on the road section under construction over a historical period.
[0071] Furthermore, according to ,get Risk Contribution Index of Abnormal Driving Behavior at Any Time .in, for Deviation from abnormal driving behavior at any given moment; , .
[0072] S106. The road risk sensitivity based on ground source information and the abnormal driving behavior risk contribution index based on air source information are integrated to obtain the comprehensive traffic risk judgment result of the construction operation section.
[0073] Specifically, it is necessary to assess the sensitivity of road risks based on geospatial information. and the risk contribution index of abnormal driving behavior based on air source information To merge: according to ,get Traffic risk assessment results at any time .in, In order to be in The road risk sensitivity obtained based on local source information at any given time; In order to be in Risk contribution index of abnormal driving behavior obtained based on spatial source information at any given time; , .
[0074] In addition, this application also provides a traffic risk assessment device for highway construction sections, such as... Figure 2 As shown, the traffic risk assessment device 200 for highway construction sections specifically includes: At least one processor 201; and a memory 202 communicatively connected to the at least one processor 201; wherein the memory 202 stores instructions executable by the at least one processor 201 to enable the at least one processor 201 to execute: Based on the network distance attenuation coefficient and the shortest path betweenness sharing rate in the transportation network, the correlation between the construction operation section and the remaining supplementary section in the expressway network is calculated to obtain the correlation neighborhood of the construction operation section. The fluctuation transmission of traffic flow in the neighboring road segments in the relevant neighborhood is calculated to obtain the neighborhood traffic flow fluctuation transmission coefficient. Based on the neighborhood traffic fluctuation transmission coefficient, the enhanced risk sensitivity coefficient of the construction section is normalized under the relevant lane capacity to obtain the road risk sensitivity of the construction section. By using drone aerial photography, based on the traffic parameters corresponding to the non-construction and construction phases, the deviation of traffic parameters for the same road segment under construction and non-construction conditions is calculated, and the deviation of abnormal driving behavior is obtained. The abnormal driving behavior deviation degree and the abnormal driving behavior influence range coefficient index of traffic accidents caused by abnormal driving behavior are used to calculate the abnormal driving behavior risk contribution index and determine the abnormal driving behavior risk contribution index. By integrating the road risk sensitivity based on ground source information with the abnormal driving behavior risk contribution index based on air source information, a comprehensive judgment result of traffic risk for the construction section is obtained.
[0075] This application's embodiments utilize traffic flow data obtained from ground-based ETC detectors to conduct macro-level traffic risk analysis. This macro-level analysis reflects the sensitivity of traffic risk in construction sections with reduced capacity to fluctuations in traffic flow from neighboring sections. Simultaneously, micro-level traffic risk analysis is conducted using aerial video footage from drones over the construction area. This micro-level analysis reflects the adverse impact of abnormal driving behavior caused by road construction on traffic risk. Finally, the macro- and micro-level analysis results are integrated to obtain a comprehensive assessment of the traffic risk of the construction section. The method described in this patent provides a more comprehensive characterization of traffic risks in highway construction sections, offering important theoretical guidance and data support for optimizing highway operation and management and ensuring the safety of construction workers.
[0076] The various embodiments in this application are described in a progressive manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the device embodiments are basically similar to the method embodiments, so the description is relatively simple; relevant parts can be referred to the descriptions of the method embodiments.
[0077] The foregoing has described specific embodiments of this application. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims may be performed in a different order than that shown in the embodiments and may still achieve the desired results. Furthermore, the processes depicted in the drawings do not necessarily require the specific or sequential order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.
[0078] The above description is merely an embodiment of this application and is not intended to limit this application. For those skilled in the art, various modifications and variations can be made to the embodiments of this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principle of the embodiments of this application should be included within the scope of the claims of this application.
Claims
1. A method for assessing traffic risks on highway construction sections, characterized in that, The method includes: Based on the network distance attenuation coefficient and the shortest path betweenness sharing rate in the traffic network, the correlation between the construction operation section and the remaining supplementary section in the highway network is calculated to obtain the correlation neighborhood of the construction operation section. The fluctuation transmission of traffic flow in the neighboring road segments of the aforementioned correlation neighborhood is calculated to obtain the neighborhood traffic flow fluctuation transmission coefficient. Based on the neighborhood traffic fluctuation transmission coefficient, the enhanced risk sensitivity coefficient of the construction section under the relevant lane capacity is normalized to obtain the road risk sensitivity of the construction section. By using drone aerial photography, based on the traffic parameters corresponding to the non-construction and construction phases, the deviation of traffic parameters for the same road segment under construction and non-construction conditions is calculated, and the deviation of abnormal driving behavior is obtained. The abnormal driving behavior deviation degree and the abnormal driving behavior influence range coefficient index of the traffic accident caused by the abnormal driving behavior are used to calculate the abnormal driving behavior risk contribution index and determine the abnormal driving behavior risk contribution index. The road risk sensitivity based on ground source information and the abnormal driving behavior risk contribution index based on air source information are fused to obtain the comprehensive traffic risk judgment result of the construction section.
2. The method for assessing traffic risks on highway construction sections according to claim 1, characterized in that, Before calculating the correlation between the construction section and the remaining supplementary sections in the highway network based on the network distance attenuation coefficient and the shortest path betweenness sharing rate in the traffic network, and obtaining the correlation neighborhood of the construction section, the method further includes: The expressway network is determined based on the connections between the network nodes; wherein, the network nodes include at least: toll stations and interchanges; The set of road segments in the expressway network is divided into a set of road segments under construction and a set of road segments not under construction. according to The network distance attenuation coefficient is obtained. ;in, Indicates the construction work section and the remaining supplementary road segments Topological distance in a road network; It is a mathematical constant; according to The shortest path betweenness sharing rate is obtained. ;in, This indicates that the highway network includes the road sections where construction work is being carried out. The number of shortest paths; This indicates that the remaining supplementary road segments are included in the highway network. The number of shortest paths; This indicates that the highway network also includes the aforementioned construction sections. and the remaining supplementary road segments The number of shortest paths; the shortest path betweenness sharing rate reflects the number of shortest paths in the construction section. and the remaining supplementary road segments The co-occurrence frequency in the shortest path of the road network, reflecting the strength of the indirect correlation between the two.
3. The method for traffic risk assessment of highway construction sections according to claim 2, characterized in that, Based on the network distance attenuation coefficient and shortest path betweenness sharing rate in the transportation network, the correlation between the construction operation section and the remaining supplementary sections in the highway network is calculated to obtain the correlation neighborhood of the construction operation section, specifically including: according to The construction section was obtained. With the remaining supplementary road segments exist The passing of time Traffic time-series correlation within hours ;in, Indicates the construction work section exist Flow rate at any moment ; Indicates the construction work section in the past The average flow rate per hour; Indicates the remaining supplementary road segment exist Flow rate at any moment ; Indicates the remaining supplementary collection section in the past The average flow rate per hour; =24 ; according to The construction section was obtained. With the remaining supplementary road segments exist Time-related correlation ;in, , , ; This is the network distance attenuation coefficient; The shortest path betweenness sharing rate; according to ,get Set of correlation indicators at different times ;in, The construction section is the section of road where the work is being carried out. and the remaining supplementary road segments exist Relevance of time; For the set of correlation indicators The elements are sorted from largest to smallest to obtain the construction section. exist ordered sequence of time ; according to , get in Correlation neighborhood at time step ;in, K The ordered sequence The front of the middle K The largest element; x For the remaining supplementary road segments Any section of the road.
4. The method for assessing traffic risks on highway construction sections according to claim 1, characterized in that, The fluctuation transmission of traffic flow in the neighboring road segments within the aforementioned correlation neighborhood is calculated to obtain the neighborhood traffic flow fluctuation transmission coefficient, specifically including: according to ,get Neighborhood flow fluctuation propagation coefficient at time 1 ;in, K For ordered sequences The front of the middle K The largest element; In the relevance neighborhood, the first The adjacent road section The flow of time, and ; In the relevance neighborhood, the first The adjacent road section in the past The average flow rate per hour.
5. The method for traffic risk assessment of highway construction sections according to claim 1, characterized in that, Based on the neighborhood traffic fluctuation transmission coefficient, the enhanced risk sensitivity coefficient of the construction section under the relevant lane capacity is normalized to obtain the road risk sensitivity of the construction section, specifically including: according to The construction section was obtained. The sensitivity coefficient of the enhanced sensitivity to risks at all times ;in, Indicates the construction work section The original traffic capacity before construction. The construction section is the section of road where the work is being carried out. Road traffic capacity during construction; The construction section exist The neighborhood flow fluctuation propagation coefficient at time t; The construction section The basic risk sensitivity coefficient, and the basic risk sensitivity coefficient This reflects road sections under construction even when they are not currently under construction. The sensitivity of traffic risk between historical traffic flow data and historical accident rate per unit of vehicle flow; according to The construction section was obtained. exist The road risk sensitivity at any given time ;in, ,and This represents the maximum value of the set of sensitivity coefficients for enhanced sensitive risks corresponding to all construction operation sections.
6. The method for assessing traffic risks on highway construction sections according to claim 1, characterized in that, Before calculating the deviation of traffic parameters for the same road segment under construction and non-construction conditions based on traffic parameters corresponding to the non-construction and construction phases using aerial photography from a drone, and before determining the deviation of abnormal driving behavior, the method further includes: Using the YOLO+DeepSort deep learning model and the LaneNet deep learning model, we can identify and locate vehicle movement trajectories and highway lane lines in aerial videos captured by UAVs from above. according to The lane keeping deviation was obtained. ;in, For highway lane width, The distance by which the vehicle's centerline deviates from the centerline of its lane. Obtain and calculate non-construction traffic parameters for traffic flow on non-construction road sections during the non-construction phase; wherein, the non-construction traffic parameters include: non-construction average speed, non-construction average acceleration, non-construction average non-construction headway, and non-construction average lane keeping deviation. Acquire and calculate the construction traffic parameters of the road section under construction during the construction phase; wherein, the construction traffic parameters include: average construction speed, average construction acceleration, average construction headway, and average construction lane keeping deviation; The non-construction phase and the construction phase refer to different stages of the same road segment.
7. The method for assessing traffic risks on highway construction sections according to claim 6, characterized in that, By using drone aerial photography, and based on traffic parameters corresponding to the non-construction and construction phases, the deviation of traffic parameters for the same road segment under construction and non-construction conditions is calculated, and the deviation of abnormal driving behavior is specifically derived, including: according to ,get Speed deviation at time ;in, for The average construction speed under the specified construction operation conditions at any given time; The non-construction average speed under the non-construction operation state; according to ,get acceleration deviation at time ;in, for The average acceleration during construction under the specified construction operation conditions at any given time; The non-construction average acceleration is the acceleration under the non-construction operation state. according to ,get Camber deviation at any given moment ;in, for The average distance between construction vehicles under the specified construction operation conditions at any given time; The distance between the non-construction vehicle heads in the non-construction operation state; according to ,get Lane keeping deviation at any time ;in, for The average lane keeping deviation during construction operations at the specified time; The non-construction average lane maintenance deviation under the non-construction operation state; according to ,get Deviation of the abnormal driving behavior at any given time ;in, , , , .
8. The method for traffic risk assessment of highway construction sections according to claim 1, characterized in that, The deviation degree of the abnormal driving behavior and the coefficient of the impact range of the abnormal driving behavior in traffic accidents caused by the abnormal driving behavior are used to calculate the behavioral risk contribution index, which determines the abnormal driving behavior risk contribution index, specifically including: according to ,get The influence range coefficient index of the abnormal driving behavior at the given time. ;in, This represents the average vehicle density of traffic flow on a road section under non-construction conditions over a historical period. This indicates the average vehicle density of traffic flow on the road section under construction over a historical period. according to ,get Risk contribution index of abnormal driving behavior at the specified time. ;in, for Deviation of the aforementioned abnormal driving behavior at any given time; , .
9. The method for assessing traffic risks on highway construction sections according to claim 1, characterized in that, The road risk sensitivity based on ground source information and the abnormal driving behavior risk contribution index based on air source information are fused to obtain the comprehensive traffic risk assessment result for the construction section, specifically including: according to ,get The comprehensive traffic risk assessment results at the current time ;in, In order to be in The road risk sensitivity obtained based on geospatial information at any given time; In order to be in The risk contribution index of the abnormal driving behavior obtained based on the spatial source information at any given time; , .
10. A traffic risk assessment device for highway construction sections, characterized in that, The 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, enabling the at least one processor to execute a traffic risk assessment method for highway construction sections according to any one of claims 1-9.