High-speed construction area mixed traffic toughness dynamic evaluation method and system

Abstract

The invention discloses a high-speed construction area mixed traffic toughness dynamic evaluation method and system, and relates to the technical field of traffic control, and the method comprises the steps: dividing a high-speed construction area mixed traffic area, obtaining M key areas, and deploying a thunder-vision integrated monitoring device; m pieces of mixed traffic flow data are collected, and a radar and vision parallel processing line is built; performing feature analysis on the data to obtain M mixed traffic association feature sets; a mixed traffic toughness evaluation system is constructed, index training is carried out, and a dynamic evaluation channel is established; and carrying out toughness evaluation on the feature set by using the channel to obtain evaluation information of the M regions, and carrying out integrated analysis to determine an overall toughness evaluation result. Therefore, the technical effects of improving the monitoring data quality, accurately quantifying the toughness of the construction area and improving the construction management level are achieved.

Description

Technical Field

[0001] The present invention relates to the field of traffic control technology, and in particular to a method and system for dynamically evaluating mixed traffic resilience in a high-speed construction zone. Background Art

[0002] In highway construction zones, managing mixed traffic flows is crucial for ensuring construction safety and smooth traffic flow. Existing traffic assessment methods primarily focus on the impact of traffic flow and construction activities, but lack a dynamic assessment of mixed traffic resilience (i.e., the system's ability to recover in the face of disturbances). Furthermore, most rely on single traffic flow monitoring devices, such as simple radar speed guns or fixed-position cameras. This has limited coverage of the mixed traffic areas within the target highway construction zone and fails to account for complex factors such as traffic flow, construction activities, topology, and accident risks. The data types collected are limited, making it difficult to fully reflect the complex characteristics of mixed traffic flows, leading to significant limitations in traffic resilience assessment. Summary of the Invention

[0003] The present invention provides a method and system for dynamically evaluating the resilience of mixed traffic in high-speed construction zones, so as to solve the technical problems in the prior art that monitoring layout blindly affects data validity and lacks quantitative evaluation of the resilience of construction zones, thereby achieving the technical effects of improving the quality of monitoring data, accurately quantifying the resilience of construction zones, and improving the level of construction management.

[0004] In a first aspect, the present invention provides a method for dynamically evaluating the mixed traffic resilience in a highway construction zone, wherein the method comprises: The mixed traffic area of ​​the target highway construction zone is divided into key areas to obtain M mixed traffic areas, and integrated radar and visual monitoring equipment are deployed in the M mixed traffic areas in turn.

[0005] M mixed traffic flow data are collected and acquired through the integrated radar and visual monitoring equipment, and a radar and visual processing dual circuit is established. The radar and visual processing dual circuit includes a radar data processing circuit and a visual data processing circuit.

[0006] The M mixed traffic flow data are analyzed in parallel using the dual-line radar processing to obtain M mixed traffic correlation feature sets.

[0007] Build a hybrid traffic resilience assessment system, conduct indicator assessment training based on the hybrid traffic resilience assessment system, and construct a hybrid traffic resilience dynamic assessment channel.

[0008] The mixed traffic resilience dynamic assessment channel is used to perform resilience assessment on the M mixed traffic associated feature sets to obtain M mixed traffic resilience assessment information, and the M mixed traffic resilience assessment information is integrated and analyzed to determine the target mixed traffic resilience assessment result.

[0009] In a feasible implementation, obtaining M mixed traffic areas includes: A traffic area division element set is preset, and the traffic area division element set includes traffic flow, construction activities, topological structure and accident risk.

[0010] A historical mixed traffic element dataset of the mixed traffic area of ​​the target highway construction zone is acquired based on the traffic area division element set.

[0011] The historical mixed traffic element dataset is used to perform element evaluation on the mixed traffic area of ​​the target highway construction zone to obtain a mixed traffic area division element parameter set.

[0012] Based on the mixed traffic area division element parameter set, regional cluster analysis and key area division are performed to obtain M mixed traffic areas.

[0013] In a feasible implementation, obtaining a mixed traffic area division element parameter set includes: According to the mixed traffic resilience assessment requirements, a regional division grid is set, and the regional division grid includes a grid shape and a grid size.

[0014] The mixed traffic area of ​​the target highway construction zone is divided into grids according to the area division grids to obtain a mixed traffic grid area.

[0015] Data mapping is performed on the mixed traffic grid area based on the historical mixed traffic element dataset to obtain a grid area traffic element dataset.

[0016] The grid area traffic element dataset is used to perform element evaluation and identification on each grid area in the mixed traffic grid area to obtain the mixed traffic area division element parameter set.

[0017] In a feasible implementation, obtaining M mixed traffic areas includes: The resilience impact assessment and division reference sorting of each element information in the traffic area division element set are performed to obtain a traffic area division element reference sequence.

[0018] According to the traffic area division element reference sequence, the division element weight factors are determined.

[0019] Regional cluster analysis is performed based on the mixed traffic area division element parameter set to obtain traffic area cluster analysis results.

[0020] The traffic area cluster analysis result is evaluated and divided into key areas according to the division factor, so as to obtain the M mixed traffic areas.

[0021] In a feasible implementation, the M mixed traffic-related feature sets are obtained, including: The M mixed traffic flow data are mapped to the radar and visual processing dual lines for data synchronization analysis to obtain M mixed traffic radar data and M mixed traffic visual data.

[0022] Vehicle trajectory tracking is performed on the M mixed traffic radar data based on the radar data processing circuit to determine M traffic vehicle operation feature sets.

[0023] The visual data processing circuit is used to perform vehicle identification and classification on the M mixed traffic visual data to obtain M traffic vehicle behavior feature sets.

[0024] The M traffic vehicle operation feature sets and the M traffic vehicle behavior feature sets are aligned and fused to obtain the M mixed traffic association feature sets.

[0025] In a feasible implementation, determining M traffic vehicle operation feature sets includes: According to the radar data processing circuit, a radar target detection network is trained and called.

[0026] The radar target detection network is used to perform vehicle target detection and identification code assignment on the M mixed traffic radar data, and a target coded vehicle set is output.

[0027] Initialize a Kalman filter, and predict and track the target coded vehicle set based on the Kalman filter to obtain a target vehicle trajectory set.

[0028] Correlated operation features are extracted based on the target vehicle trajectory set to determine M traffic vehicle operation feature sets, where the M traffic vehicle operation feature sets include traffic flow, traffic speed, and traffic density.

[0029] In a feasible implementation, the construction of a dynamic assessment channel for hybrid traffic resilience includes: The hybrid traffic resilience assessment system is subjected to correlation index decomposition and feature data mining to obtain a hybrid traffic historical resilience index dataset.

[0030] A deep neural network structure is used to perform resilience assessment training on the mixed traffic history resilience index dataset to obtain a resilience index evaluation branch channel.

[0031] A decision impact analysis is performed on the toughness index evaluation branch channel to determine the toughness index channel decision weight factor.

[0032] Based on the resilience index channel decision weight factor, the resilience index evaluation branch channels are weighted and combined in parallel to construct the mixed traffic resilience dynamic evaluation channel.

[0033] In a feasible implementation, obtaining a mixed traffic historical resilience index dataset includes: The hybrid traffic resilience assessment system is decomposed into related indicators to obtain a set of hybrid traffic resilience related assessment indicators.

[0034] A toughness indicator association depth set is determined according to the toughness indicator channel decision weight factor.

[0035] Traffic feature data mining is performed based on the resilience index association depth set to obtain the mixed traffic historical resilience index data set.

[0036] In a feasible implementation, determining the target mixed traffic resilience assessment result includes: The resilience impact range of each traffic area in the M mixed traffic areas is evaluated to obtain the resilience impact coefficients of the M traffic areas.

[0037] The M mixed traffic resilience assessment information is integrated and analyzed based on the M traffic area resilience influence coefficients to determine the target mixed traffic resilience assessment result.

[0038] In a second aspect, the present invention further provides a dynamic assessment system for mixed traffic resilience in a highway construction zone, wherein the dynamic assessment system for mixed traffic resilience in a highway construction zone comprises: The area division and monitoring deployment module is used to divide the mixed traffic area of ​​the target highway construction area into key areas, obtain M mixed traffic areas, and deploy integrated radar and visual monitoring equipment in the M mixed traffic areas in turn.

[0039] The data acquisition module is used to collect M mixed traffic flow data through the integrated radar and visual monitoring equipment, and build a radar and visual processing dual circuit, which includes a radar data processing circuit and a visual data processing circuit.

[0040] The feature analysis module is used to perform feature analysis on the M mixed traffic flow data in parallel using the radar visual processing dual lines to obtain M mixed traffic related feature sets.

[0041] The assessment capacity building module is used to build a hybrid traffic resilience assessment system, conduct indicator assessment training based on the hybrid traffic resilience assessment system, and construct a hybrid traffic resilience dynamic assessment channel.

[0042] An evaluation and result integration module is used to perform resilience evaluation on the M mixed traffic associated feature sets using the mixed traffic resilience dynamic evaluation channel to obtain M mixed traffic resilience evaluation information, integrate and analyze the M mixed traffic resilience evaluation information, and determine the target mixed traffic resilience evaluation result.

[0043] The present invention discloses a method and system for dynamically evaluating the resilience of mixed traffic in a highway construction zone, comprising: first, dividing the mixed traffic area of ​​a target highway construction zone into key areas to obtain M mixed traffic areas, and sequentially deploying radar-visual integrated monitoring equipment in these areas; then, collecting M mixed traffic flow data through the radar-visual integrated monitoring equipment, and building a radar-visual processing dual line, including a radar data processing line and a visual data processing line; then, performing feature analysis on the M mixed traffic flow data in parallel using the radar-visual processing dual line to obtain M mixed traffic related feature sets; thereafter, building a mixed traffic resilience evaluation system, and performing indicator evaluation training based on the system to construct a dynamic evaluation channel for mixed traffic resilience; finally, using the evaluation channel to perform resilience evaluation on the M mixed traffic related feature sets to obtain M mixed traffic resilience evaluation information, and integrating and analyzing the evaluation information to ultimately determine the target mixed traffic resilience evaluation result. The method and system for dynamically evaluating the resilience of mixed traffic in a highway construction zone disclosed by the present invention solve the technical problems that the monitoring layout blindly affects the data validity and lacks quantitative evaluation of the resilience of the construction zone, and achieves the technical effects of improving the quality of monitoring data, accurately quantifying the resilience of the construction zone, and improving the level of construction management. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 This is a flow chart of a method for dynamically evaluating mixed traffic resilience in a high-speed construction zone according to the present invention.

[0045] Figure 2 This is a structural schematic diagram of a dynamic assessment system for mixed traffic resilience in a high-speed construction zone according to the present invention.

[0046] Explanation of the reference numerals: regional division and monitoring deployment module 11, data collection module 12, feature analysis module 13, assessment capacity building module 14, assessment and result integration module 15. DETAILED DESCRIPTION

[0047] The above technical solution will be described in detail below in conjunction with the accompanying drawings and specific implementation methods of the specification to better understand the above technical solution. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments of the present invention. It should be understood that the present invention is not limited to the example embodiments used only to explain the present invention. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention. In addition, it should be noted that, for the convenience of description, only the parts related to the present invention, rather than all, are shown in the drawings.

[0048] Example 1, as Figure 1 The figure is a flow chart of a method for dynamically evaluating the mixed traffic resilience in a highway construction zone according to the present invention. The method comprises: S100: Divide the mixed traffic area of ​​the target highway construction zone into key areas to obtain M mixed traffic areas, and sequentially deploy integrated radar and visual monitoring equipment in the M mixed traffic areas.

[0049] Specifically, based on predetermined traffic elements and assessment requirements, the mixed traffic area within the highway construction zone is divided into several sub-areas with varying traffic characteristics and importance, such as those with high traffic volume, frequent construction activities, high accident risk, or complex topology. The integrated radar and visual monitoring equipment combines radar and visual sensors, capable of simultaneously collecting vehicle radar signals (such as speed and distance) and visual images (such as vehicle type and color).

[0050] Through the division of key areas and the deployment of integrated radar and vision monitoring equipment, key areas of mixed traffic in highway construction zones can be accurately located, avoiding blind monitoring deployment.

[0051] In some embodiments, obtaining M mixed traffic areas includes: A traffic area division element set is preset, and the traffic area division element set includes traffic flow, construction activities, topological structure and accident risk; based on the traffic area division element set, a historical mixed traffic element dataset of the mixed traffic area of ​​the target highway construction area is collected and obtained; the historical mixed traffic element dataset is used to perform element evaluation on the mixed traffic area of ​​the target highway construction area to obtain a mixed traffic area division element parameter set; regional clustering analysis and key area division are performed based on the mixed traffic area division element parameter set to obtain M mixed traffic areas.

[0052] Specifically, the traffic zone delineation feature set includes a collection of core indicators used to analyze and zonate mixed traffic within highway construction zones. These elements reflect the complexity and importance of traffic zones. Preferred indicators include traffic flow (such as vehicle throughput, vehicle type ratio, and flow rate), construction activity (construction time, spatial extent, and intensity), topology (road branch connectivity, number of lanes, turning channels, etc.), and accident risk (historical accident frequency, accident density, or probability of high-risk events).

[0053] Specifically, the historical mixed traffic element dataset refers to long-term series data covering the above-mentioned division elements and collected from the target highway construction area, where the data acquisition sources include but are not limited to structured data collected by traffic management platforms, sensors, drones, or monitoring.

[0054] Specifically, first, a set of traffic characteristic indicators, including flow density, high accident probability, topological complexity, etc., are preset to establish a unified evaluation standard and form a traffic area division element set; then, multi-dimensional mixed traffic data across time periods is obtained from the traffic monitoring system, traffic flow simulation platform or historical database deployed in the construction area as a historical mixed traffic element dataset to provide basic information for subsequent area division; then, the data is cleaned and analyzed, and the numerical features of each area to be divided in each element dimension are extracted. Then, an unsupervised clustering method (such as K-means) is used to divide all area samples, where the clustering dimensions may include construction intensity, flow volatility, topological connection density, potential accident risk, etc., and finally M mixed traffic areas with similar traffic characteristics are formed, and the area division results are output.

[0055] The above steps, by presetting a set of traffic zone demarcation elements and collecting a historical mixed traffic element dataset, comprehensively consider the various factors affecting traffic resilience and ensure the comprehensiveness and scientific nature of the zone demarcation. The use of historical data for element evaluation, resulting in a set of mixed traffic zone demarcation element parameters, further quantifies the characteristics of each zone and makes the zone demarcation more precise. Regional cluster analysis and key zone demarcation based on the element parameter set effectively avoids blind monitoring deployment, ensures that monitoring equipment is deployed in key areas, and thus improves the effectiveness of monitoring data.

[0056] In some implementations, obtaining a mixed traffic area division element parameter set includes: According to the requirements of mixed traffic resilience assessment, a regional division grid is set, and the regional division grid includes a grid shape and a grid size; the mixed traffic area of ​​the target highway construction zone is grid-divided according to the regional division grid to obtain a mixed traffic grid area; data mapping is performed on the mixed traffic grid area based on the historical mixed traffic element dataset to obtain a grid area traffic element dataset; the grid area traffic element dataset is used to perform element assessment identification on each grid area in the mixed traffic grid area to obtain the mixed traffic area division element parameter set.

[0057] Specifically, the area division grid is a virtual grid used to divide the mixed traffic area of ​​the target highway construction zone, that is, it is used to subdivide the large construction area into smaller units. Among them, the area division grid requires defining the grid shape (such as square, rectangle, hexagon, etc.) and grid size (such as side length, area, etc.). For example, the size can be set according to road density or traffic flow sampling interval.

[0058] Specifically, first, according to the traffic resilience assessment objectives of the construction area (such as maximizing the discovery of high-risk traffic aggregation areas), the regional division grid is set, including the grid shape (such as hexagonal shape is conducive to adjacency analysis) and grid size (such as 100m×100m or dynamically adjusted scale); then, based on the above grid shape and size, the target highway construction area is spatially divided to generate multiple mixed traffic grid areas, where each grid has a unique spatial number; further, according to the regional benchmark point of the construction area (such as the starting point of the warning area), the target highway construction area is aligned with the historical highway construction area corresponding to the historical mixed traffic element dataset, and then the data in the historical mixed traffic element dataset (such as construction frequency, traffic flow time variation, accident statistics) is mapped to the corresponding grid area according to coordinates or regional numbers to form a grid area traffic element dataset.

[0059] Furthermore, the traffic indicators within each grid are evaluated and calculated. For example, weighted scoring, risk grading, etc. can be used to assign a grade or identification label to each indicator, and the evaluation label and numerical characteristics of each grid are output to finally form a set of mixed traffic area division element parameters, which provides an input basis for subsequent clustering and zoning.

[0060] For example, the principal component analysis method can be used to determine the weighted weights of multiple factors, and then combined with the grid area traffic factor dataset, the division factor parameters of each grid area can be calculated to perform factor evaluation and identification.

[0061] Through the above process, complex traffic areas can be subdivided into smaller units, which facilitates the refined evaluation of traffic elements in each unit, thereby providing more accurate and detailed data support for subsequent regional cluster analysis and key area division.

[0062] In some implementations, obtaining M mixed traffic areas includes: Perform resilience impact assessment and division reference sorting on the information of each element in the traffic area division element set to obtain a traffic area division element reference sequence; determine the division element weight factor based on the traffic area division element reference sequence; perform regional clustering analysis based on the mixed traffic area division element parameter set to obtain traffic area clustering analysis results; perform key area assessment and division on the traffic area clustering analysis results according to the division element weight factor to obtain the M mixed traffic areas.

[0063] Specifically, the division factor weight factor is a weight value assigned to each element based on the traffic area division factor reference sequence. The weight factor reflects the importance of each element in dividing the traffic area.

[0064] Specifically, the resilience impact assessment is first conducted on each element in the traffic zone demarcation element set, analyzing its impact on the stability and response (recovery) capacity of the traffic system. The multiple elements are then ranked accordingly to form a traffic zone demarcation element reference sequence, which reflects the relative importance of each element in the demarcation process. Based on this reference sequence, a weighted ranking method, such as the analytic hierarchy process (AHP), or the entropy weight method, is then used to determine the weight factor for each demarcation element. This weight factor is used to reflect the contribution of different elements to the demarcation process in subsequent cluster analysis.

[0065] Furthermore, based on the aforementioned mixed traffic area partitioning parameter set (i.e., the numerical representation of all areas on each partitioning element), combined with the aforementioned weight factors, regional clustering analysis is performed. The corresponding traffic area clustering analysis results are output as several areas with similar traffic characteristics. These areas can be regarded as preliminary partitioning results. Optional clustering methods include K-Means, Fuzzy C-Means (FCM), DBSCAN, etc.

[0066] Furthermore, based on the cluster analysis results, a critical assessment of each cluster area is performed according to the weight factor of the division factors, that is, the quantitative evaluation value of each cluster area is weighted and calculated, and then the areas with a key impact on system resilience or traffic efficiency are identified, and M mixed traffic areas are obtained.

[0067] Through the above method, the weight of each factor in the traffic area division can be determined more scientifically, ensuring that the key areas divided are more in line with the actual needs of mixed traffic resilience assessment, and ensuring that the M mixed traffic areas obtained are highly representative and targeted, which helps to improve the quality of subsequent monitoring data and the accuracy of traffic resilience assessment.

[0068] S200: M mixed traffic flow data are collected and acquired through the integrated radar and visual monitoring equipment, and a radar and visual processing dual circuit is established, wherein the radar and visual processing dual circuit includes a radar data processing circuit and a visual data processing circuit.

[0069] Specifically, the integrated radar and visual monitoring equipment uses the heterogeneous perception capabilities of radar and visual sensors to perform fused perception processing on multiple types of traffic participants in mixed traffic flows (such as motor vehicles, non-motor vehicles, pedestrians, etc.), thereby improving the accuracy and robustness of target detection. Among them, the radar channel focuses on extracting motion characteristics (such as speed, acceleration, trajectory stability, etc.); the visual channel focuses on extracting appearance and behavioral characteristics (such as category, size, behavior pattern, etc.).

[0070] Specifically, the dual radar and visual processing circuits include a radar data processing circuit and a visual data processing circuit. The radar data processing circuit is a system or path for processing data collected by the radar, including steps such as signal amplification, filtering, target detection, and trajectory tracking to extract information such as the vehicle's speed and position. The visual data processing circuit is a system or path for processing data collected by visual sensors (such as cameras), involving steps such as image preprocessing, target detection, classification and recognition, and behavior analysis to extract information such as the vehicle's type, color, and driving behavior.

[0071] S300: Using the radar visual processing dual lines to perform feature analysis on the M mixed traffic flow data in parallel to obtain M mixed traffic associated feature sets.

[0072] The radar and visual processing dual lines process the radar signal part and the visual signal part of the M mixed traffic flow data through the radar data processing line and the visual data processing line respectively, and combine the data fusion method (such as through feature alignment and fusion mechanism) to construct a unified mixed traffic flow feature representation to obtain the final M mixed traffic correlation feature set.

[0073] For example, in a mixed traffic area, the radar channel detects multiple high-speed motor vehicle targets, and the visual channel identifies a large number of non-motor vehicles and pedestrians crossing each other. After fusion, a characteristic label of "high-speed motor vehicle-slow traffic high interaction" is formed, providing data support for subsequent regional risk identification and strategy formulation.

[0074] In some embodiments, obtaining M mixed traffic-related feature sets includes: The M mixed traffic flow data are mapped to the radar and visual processing dual lines for data synchronization analysis to obtain M mixed traffic radar data and M mixed traffic visual data; vehicle trajectories are tracked on the M mixed traffic radar data based on the radar data processing line to determine M traffic vehicle operation feature sets; vehicle identification and classification are performed on the M mixed traffic visual data using the visual data processing line to obtain M traffic vehicle behavior feature sets; the M traffic vehicle operation feature sets and the M traffic vehicle behavior feature sets are aligned and fused to obtain the M mixed traffic association feature sets.

[0075] Specifically, M mixed traffic flow data are first input into the radar and visual processing system. Parallel data analysis is performed via the radar data processing circuit and the visual data processing circuit, yielding M mixed traffic radar data and M mixed traffic visual data, respectively. The radar and visual processing system employs mechanisms such as timestamp synchronization and spatial registration to ensure consistency between radar and visual data in both temporal and spatial dimensions.

[0076] Specifically, based on the radar data processing circuit, target detection and trajectory tracking are performed on M mixed traffic radar data to extract the operating characteristics of each traffic target, including speed, acceleration, direction of travel, lane changing behavior, road section travel time, etc., thereby obtaining M traffic vehicle operation feature sets.

[0077] Simultaneously, a visual data processing circuit is used to perform target detection, recognition, and classification on M mixed traffic visual data to obtain the behavioral characteristics of traffic participants, including vehicle type (car, truck, bus, bicycle, etc.), driving behavior (following, lane changing, sudden braking, etc.), interaction behavior with pedestrians or non-motor vehicles, etc., and output them as M traffic vehicle behavior feature sets.

[0078] Furthermore, the M traffic vehicle operation feature sets and the M traffic vehicle behavior feature sets are aligned at the target level, such as based on target ID or trajectory matching (determined in combination with the time series and the positional relationship of the M mixed traffic areas), and M mixed traffic association feature sets are generated through feature fusion strategies such as feature splicing, weighted fusion, and graph neural network modeling.

[0079] The above method maps M mixed traffic flow data to radar and visual processing circuits, extracting traffic operation and behavior characteristics. Based on feature alignment and fusion strategies, M mixed traffic correlation feature sets are generated for subsequent traffic state assessment, behavior modeling, or intelligent scheduling decisions. The parallel processing of the radar and visual processing circuits improves data processing efficiency and accuracy, while ensuring the reliability and timeliness of subsequent feature analysis.

[0080] In some implementations, determining M traffic vehicle operation characteristic sets includes: According to the radar data processing circuit, a radar target detection network is trained and called; the radar target detection network is used to perform vehicle target detection and identification code assignment on the M mixed traffic radar data, and a target coded vehicle set is output; a Kalman filter is initialized, and the target coded vehicle set is predicted and tracked based on the Kalman filter to obtain a target vehicle trajectory set; and associated operation feature extraction is performed based on the target vehicle trajectory set to determine M traffic vehicle operation feature sets, wherein the M traffic vehicle operation feature sets include traffic flow, traffic speed, and traffic density.

[0081] Specifically, the radar target detection network is a pre-built or existing machine learning or deep learning-based algorithm model (such as PointNet++, YOLO-Radar, CenterPoint, etc.), which is specifically designed to detect and identify targets (such as vehicles) from radar data and output corresponding information such as the target's location and speed. Each detected vehicle target is assigned a unique identification code to enable accurate tracking and differentiation of different targets in subsequent processing. The target coded vehicle set refers to the set of vehicles detected and assigned identification codes by the radar target detection network, including the unique identification codes of multiple vehicles and related detection information (such as location, speed, etc.).

[0082] Specifically, in determining the M sets of traffic vehicle operational characteristics, a radar target detection network is first trained based on labeled radar point cloud or radar image data to achieve high-precision detection of traffic targets. The radar target detection network is then invoked to perform target detection on each frame of radar data, outputting a set of detected vehicle targets and assigning each target a unique identification code (ID) to form a set of target-coded vehicles. Next, a Kalman filter is initialized to perform state estimation and trajectory prediction for each target-coded vehicle. This involves combining historical observations with current detection results to perform target association and trajectory update.

[0083] Furthermore, the target vehicle trajectory set obtained above is used to further extract associated operation features, including traffic flow, traffic speed, traffic density, etc., to determine M traffic vehicle operation feature sets.

[0084] Through deep processing of radar data, the above process is able to extract key features that can reflect the operating status of traffic flow, providing accurate operating characteristic data support for subsequent traffic resilience assessment.

[0085] S400: Build a hybrid traffic resilience assessment system, conduct indicator assessment training based on the hybrid traffic resilience assessment system, and construct a hybrid traffic resilience dynamic assessment channel.

[0086] Specifically, the hybrid traffic resilience assessment system is a comprehensive framework for evaluating the responsiveness, recovery, and system stability of multiple traffic types (e.g., civilian vehicles, construction vehicles, and emergency vehicles) within highway construction zones when faced with disruptions (e.g., road closures and accidents). For example, the system comprises a set of quantitative indicators representing the "vulnerability-resilience-dynamic adaptability" criterion. By deconstructing and mining these resilience indicators, combined with deep neural network training and multi-channel weighting, a dynamic hybrid traffic resilience assessment channel with adaptive and dynamic response capabilities can be developed to support traffic system status assessment and strategy adjustments.

[0087] Specifically, the resilience assessment system constructed for mixed traffic scenarios includes resilience assessment indicators in multiple dimensions, such as traffic flow stability, traffic recovery capability, system redundancy, and abnormal response capability. The resilience assessment system is determined based on the needs of the target area or scenario. For example, the higher the design speed, the greater the traffic volume, and the section belongs to the trunk traffic line, the stricter the requirements of the corresponding resilience assessment system.

[0088] In some embodiments, constructing a dynamic assessment channel for hybrid traffic resilience includes: The hybrid traffic resilience assessment system is subjected to associated indicator decomposition and feature data mining to obtain a hybrid traffic historical resilience indicator data set; a deep neural network structure is used to perform resilience assessment training on the hybrid traffic historical resilience indicator data set to obtain a resilience indicator assessment branch channel; a decision impact analysis is performed on the resilience indicator assessment branch channel to determine the resilience indicator channel decision weight factor; based on the resilience indicator channel decision weight factor, the resilience indicator assessment branch channels are weightedly combined in parallel to construct the hybrid traffic resilience dynamic assessment channel.

[0089] Specifically, a dynamic evaluation channel for hybrid traffic resilience is constructed. First, based on historical hybrid traffic operation data, event response data, etc., feature data mining is carried out around each resilience indicator, and the high-level system resilience evaluation indicators in the hybrid traffic resilience evaluation system (such as traffic flow sustainability, post-accident recovery time, alternative path utilization, etc.) are decomposed into indicator hierarchies to obtain a set of related evaluation indicators for hybrid traffic resilience. This indicator set contains basic statistical indicators or combined characteristics that can affect or reflect traffic resilience, such as vehicle speed fluctuation rate, travel time variance, queue length growth rate, etc., thereby improving the computability and perceptibility of subsequent indicators.

[0090] Specifically, using the mixed traffic historical resilience index dataset as sample data, the deep neural network structure is trained for resilience assessment based on supervised learning until the trained deep neural network structure meets the preset termination conditions (such as reaching a preset number of times or the evaluation performance meets a preset threshold), and the deep neural network structure is output as a resilience index evaluation branch channel.

[0091] Specifically, by repeating the above training process, multiple resilience indicator evaluation branch channels can be obtained, which are used to independently evaluate and output different categories of resilience indicators. Then, the role of each branch channel in the overall resilience evaluation is evaluated, such as by using feature importance evaluation, attention scoring mechanism or Shapley value, to extract the contribution of each indicator channel to the final judgment result, and normalize the contribution value to obtain it as the weight of each resilience indicator evaluation branch channel.

[0092] Furthermore, after combining multiple evaluation result channels in parallel, they are linearly or nonlinearly weighted according to the corresponding weight factors, thereby generating a comprehensive, continuously output mixed traffic resilience dynamic evaluation channel to dynamically evaluate the resilience level of the traffic status in the construction area.

[0093] In some implementations, obtaining a mixed traffic historical resilience indicator dataset includes: The hybrid traffic resilience assessment system is decomposed into related indicators to obtain a set of hybrid traffic resilience related assessment indicators; the resilience indicator related depth set is determined based on the resilience indicator channel decision weight factor; and traffic feature data are mined based on the resilience indicator related depth set to obtain the hybrid traffic historical resilience indicator data set.

[0094] Specifically, first, the macro indicators of each dimension in the resilience indicator system are decomposed according to the evaluation objectives, and the basic evaluation factors that can be quantified, collected and calculated are extracted, and the resilience correlation evaluation indicator set that can be used for subsequent modeling and analysis is stored and generated; then, according to the importance of each indicator channel, the management depth of each indicator, that is, the strength of the correlation, is set, and then the correlation depth of each indicator corresponding to multiple resilience indicator channel decision weight factors is integrated, and multiple resilience correlation evaluation indicators are serialized based on the integrated integrated correlation depth, and a preset number is selected as the mixed traffic historical resilience indicator data set.

[0095] Through the above-mentioned method and steps, historical data features closely related to resilience assessment can be accurately extracted, providing high-quality data support for subsequent assessment model training and resilience analysis, and ensuring the accuracy and reliability of the assessment results.

[0096] S500: Use the mixed traffic resilience dynamic assessment channel to perform resilience assessment on the M mixed traffic associated feature sets to obtain M mixed traffic resilience assessment information, integrate and analyze the M mixed traffic resilience assessment information, and determine the target mixed traffic resilience assessment result.

[0097] In some embodiments, determining a target mixed traffic resilience assessment result includes: Perform a resilience impact range assessment on each of the M mixed traffic areas to obtain M traffic area resilience impact coefficients; perform an integrated analysis on the M mixed traffic resilience assessment information based on the M traffic area resilience impact coefficients to determine the target mixed traffic resilience assessment result.

[0098] Specifically, the traffic area resilience impact coefficient is a quantitative indicator derived from the resilience impact range assessment. It is used to measure the weight of each mixed traffic area's impact on overall traffic resilience. By combining multiple mixed traffic resilience assessment information with the corresponding resilience impact coefficient, the contribution of each area is comprehensively considered to ultimately determine the target mixed traffic resilience assessment result.

[0099] Specifically, first, M mixed traffic correlation feature sets are input one by one into the mixed traffic resilience dynamic assessment channel for resilience assessment to obtain the corresponding mixed traffic resilience assessment information; then, the resilience impact range of the M mixed traffic areas is evaluated respectively, such as determining the degree of influence of each area on the overall traffic resilience based on the length, number of lanes, speed limit, safe vehicle distance, etc. of different mixed traffic areas, and obtaining the traffic area resilience impact coefficient of each area.

[0100] Furthermore, the M mixed traffic resilience assessment information is integrated and analyzed, i.e., the assessment information of each area is weighted and integrated to ultimately determine the target mixed traffic resilience assessment result. This comprehensively considers the characteristics of each mixed traffic area and its impact on overall resilience, obtaining a comprehensive and accurate assessment result.

[0101] In the above method flow, the introduction of resilience impact range assessment and traffic area resilience impact coefficient enables the integrated analysis to highlight the impact of key areas, avoid the simple averaging of assessment information from each region, and thus more scientifically reflect the overall traffic resilience status.

[0102] In summary, the method for dynamic assessment of mixed traffic resilience in a highway construction zone provided by the present invention has the following technical effects: First, the mixed traffic area of ​​the target highway construction zone is divided into key areas to obtain M mixed traffic areas, and integrated radar and visual monitoring equipment is deployed in these areas in turn; then, M mixed traffic flow data are collected through the integrated radar and visual monitoring equipment, and a radar and visual processing dual line is built, including a radar data processing line and a visual data processing line; then, the M mixed traffic flow data are analyzed in parallel using the radar and visual processing dual lines to obtain M mixed traffic related feature sets; then, a mixed traffic resilience assessment system is built, and indicator evaluation training is carried out based on the system to construct a mixed traffic resilience dynamic assessment channel; finally, the assessment channel is used to perform resilience assessment on the M mixed traffic related feature sets to obtain M mixed traffic resilience assessment information, and these assessment information are integrated and analyzed to finally determine the target mixed traffic resilience assessment results, thereby achieving the technical effects of improving the quality of monitoring data, accurately quantifying the resilience of the construction area, and improving the level of construction management.

[0103] Example 2, as Figure 2 This is a schematic diagram of the structure of a dynamic assessment system for mixed traffic resilience in a high-speed construction zone according to the present invention. Figure 1 The flow chart of the dynamic assessment method of mixed traffic resilience in a high-speed construction zone of the present invention can be shown as follows: Figure 2 The structure shown is implemented.

[0104] Based on the same concept as the method for dynamically evaluating the mixed traffic resilience in a highway construction zone described in the embodiment, the present invention further provides a system for dynamically evaluating the mixed traffic resilience in a highway construction zone, comprising: The area division and monitoring deployment module 11 is used to divide the mixed traffic area of ​​the target highway construction area into key areas, obtain M mixed traffic areas, and deploy integrated radar and visual monitoring equipment in the M mixed traffic areas in turn.

[0105] The data acquisition module 12 is used to collect M mixed traffic flow data through the integrated radar and visual monitoring equipment, and build a radar and visual processing dual circuit, which includes a radar data processing circuit and a visual data processing circuit.

[0106] The feature analysis module 13 is used to perform feature analysis on the M mixed traffic flow data in parallel using the radar visual processing dual lines to obtain M mixed traffic related feature sets.

[0107] The assessment capability building module 14 is used to build a hybrid traffic resilience assessment system, conduct indicator assessment training based on the hybrid traffic resilience assessment system, and construct a hybrid traffic resilience dynamic assessment channel.

[0108] The evaluation and result integration module 15 is used to use the mixed traffic resilience dynamic evaluation channel to perform resilience evaluation on the M mixed traffic associated feature sets, obtain M mixed traffic resilience evaluation information, integrate and analyze the M mixed traffic resilience evaluation information, and determine the target mixed traffic resilience evaluation result.

[0109] In some embodiments, the area division and monitoring deployment module 11 includes: The element preset unit is used to preset a traffic area division element set, wherein the traffic area division element set includes traffic flow, construction activities, topological structure and accident risk.

[0110] The data collection unit is used to collect and obtain a historical mixed traffic element dataset of the mixed traffic area of ​​the target highway construction zone based on the traffic area division element set.

[0111] An element evaluation unit is used to use the historical mixed traffic element data set to perform element evaluation on the mixed traffic area of ​​the target highway construction zone to obtain a mixed traffic area division element parameter set.

[0112] The area division unit is used to perform regional cluster analysis and key area division based on the mixed traffic area division element parameter set to obtain M mixed traffic areas.

[0113] In some implementations, the factor evaluation unit in the area division and monitoring deployment module 11 includes: The grid setting unit is used to set the regional division grid according to the mixed traffic resilience assessment requirements, and the regional division grid includes a grid shape and a grid size.

[0114] The grid division unit is used to divide the mixed traffic area of ​​the target highway construction area into grids according to the area division grids to obtain a mixed traffic grid area.

[0115] A data mapping unit is used to perform data mapping on the mixed traffic grid area based on the historical mixed traffic element dataset to obtain a grid area traffic element dataset.

[0116] The element evaluation and identification unit is used to use the grid area traffic element data set to perform element evaluation and identification on each grid area in the mixed traffic grid area to obtain the mixed traffic area division element parameter set.

[0117] In some implementations, the region division unit in the region division and monitoring deployment module 11 includes: The resilience impact assessment and ranking unit is used to perform resilience impact assessment and division reference ranking on the information of each element in the traffic area division element set to obtain a traffic area division element reference sequence.

[0118] The division element weight factor determination unit is used to determine the division element weight factor according to the traffic area division element reference sequence.

[0119] The regional cluster analysis unit is used to perform regional cluster analysis based on the mixed traffic area division element parameter set to obtain traffic area cluster analysis results.

[0120] The key area evaluation and division unit is used to perform key area evaluation and division on the traffic area cluster analysis result according to the division factor weight factor to obtain the M mixed traffic areas.

[0121] In some embodiments, the feature analysis module 13 includes: The data synchronization and analysis unit is used to map the M mixed traffic flow data to the radar and visual processing dual lines for data synchronization and analysis to obtain M mixed traffic radar data and M mixed traffic visual data.

[0122] A vehicle trajectory tracking unit is used to track vehicle trajectories on the M mixed traffic radar data based on the radar data processing circuit to determine M traffic vehicle operation feature sets.

[0123] The vehicle identification and classification unit is used to perform vehicle identification and classification on the M mixed traffic visual data using the visual data processing circuit to obtain M traffic vehicle behavior feature sets.

[0124] The feature set alignment and fusion unit is used to align and fuse the M traffic vehicle operation feature sets and the M traffic vehicle behavior feature sets to obtain the M mixed traffic association feature sets.

[0125] In some implementations, the vehicle trajectory tracking unit in the feature analysis module 13 includes: The radar target detection network training and calling unit is used to train and call the radar target detection network according to the radar data processing circuit.

[0126] The vehicle target detection and identification code allocation unit is used to use the radar target detection network to perform vehicle target detection and identification code allocation on the M mixed traffic radar data, and output a target coded vehicle set.

[0127] The Kalman filter initialization and prediction tracking unit is used to initialize the Kalman filter, predict and track the target coded vehicle set based on the Kalman filter, and obtain the target vehicle trajectory set.

[0128] The associated operation feature extraction unit is used to extract associated operation features based on the target vehicle trajectory set to determine M traffic vehicle operation feature sets, wherein the M traffic vehicle operation feature sets include traffic flow, traffic speed and traffic density.

[0129] In some embodiments, the assessment capability building module 14 includes: The correlation index decomposition and data mining unit is used to perform correlation index decomposition and feature data mining on the hybrid traffic resilience assessment system to obtain a hybrid traffic historical resilience index data set.

[0130] The resilience assessment training unit is used to use a deep neural network structure to perform resilience assessment training on the mixed traffic history resilience index data set to obtain a resilience index assessment branch channel.

[0131] The decision impact analysis unit is used to perform decision impact analysis on the resilience index evaluation branch channel and determine the resilience index channel decision weight factor.

[0132] A dynamic evaluation channel construction unit is used to perform parallel weighted combination of the resilience index evaluation branch channels based on the resilience index channel decision weight factor to construct the mixed traffic resilience dynamic evaluation channel.

[0133] In some implementations, the correlation indicator decomposition and data mining unit in the assessment capability building module 14 includes: The correlation indicator decomposition unit is used to decompose the correlation indicators of the hybrid traffic resilience evaluation system to obtain a set of hybrid traffic resilience correlation evaluation indicators.

[0134] The toughness indicator associated depth set determining unit is used to determine the toughness indicator associated depth set according to the toughness indicator channel decision weight factor.

[0135] The traffic feature data mining unit is used to perform traffic feature data mining based on the resilience indicator association depth set to obtain the mixed traffic history resilience indicator data set.

[0136] In some embodiments, the evaluation and results integration module 15 includes: The resilience influence range evaluation unit is used to evaluate the resilience influence range of each traffic area in the M mixed traffic areas to obtain the resilience influence coefficients of the M traffic areas.

[0137] An evaluation information integration and result determination unit is used to integrate and analyze the M mixed traffic resilience evaluation information based on the M traffic area resilience influence coefficients to determine the target mixed traffic resilience evaluation result.

[0138] It should be understood that the embodiments mentioned in this specification focus on their differences from other embodiments. The specific embodiments in the aforementioned embodiment one are also applicable to a dynamic assessment system for mixed traffic resilience in a high-speed construction area described in embodiment two. For the sake of brevity of the specification, no further elaboration will be given here.

[0139] It should be understood that the embodiments disclosed in the present invention and the above description can enable those skilled in the art to use the present invention to implement the present invention. At the same time, the present invention is not limited to the embodiments mentioned above. It should be understood that those skilled in the art can still modify the technical solutions described in the above embodiments or replace some of the technical features therein with equivalents; and such modifications or replacements do not deviate from the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention and are all included in the scope of protection of the present invention.

Claims

1. A dynamic assessment method for mixed traffic resilience in a highway construction zone, characterized by: include: Divide the mixed traffic area of ​​the target highway construction zone into key areas to obtain M mixed traffic areas, and sequentially deploy integrated radar and visual monitoring equipment in the M mixed traffic areas; The radar and visual integrated monitoring equipment is used to collect M mixed traffic flow data and establish a radar and visual processing dual circuit, wherein the radar and visual processing dual circuit includes a radar data processing circuit and a visual data processing circuit; Using the radar visual processing dual lines to perform feature analysis on the M mixed traffic flow data in parallel to obtain M mixed traffic correlation feature sets; Establish a hybrid traffic resilience assessment system, conduct indicator assessment training based on the hybrid traffic resilience assessment system, and build a dynamic assessment channel for hybrid traffic resilience; The mixed traffic resilience dynamic assessment channel is used to perform resilience assessment on the M mixed traffic associated feature sets to obtain M mixed traffic resilience assessment information, and the M mixed traffic resilience assessment information is integrated and analyzed to determine the target mixed traffic resilience assessment result.

2. A dynamic assessment method for mixed traffic resilience in a high-speed construction zone according to claim 1, characterized in that: The obtained M mixed traffic areas include: A traffic area division element set is preset, wherein the traffic area division element set includes traffic flow, construction activities, topological structure, and accident risk; Acquire a historical mixed traffic element dataset of the mixed traffic area of ​​the target highway construction zone based on the traffic area division element set; Using the historical mixed traffic element dataset to perform element evaluation on the mixed traffic area of ​​the target highway construction zone, and obtaining a mixed traffic area division element parameter set; Based on the mixed traffic area division element parameter set, regional cluster analysis and key area division are performed to obtain M mixed traffic areas.

3. The method for dynamic assessment of mixed traffic resilience in a high-speed construction zone according to claim 2, characterized in that: The step of obtaining the mixed traffic area division element parameter set includes: According to the mixed traffic resilience assessment requirements, a regional division grid is set, wherein the regional division grid includes a grid shape and a grid size; Dividing the mixed traffic area of ​​the target highway construction zone into a grid according to the regional division grid to obtain a mixed traffic grid area; Performing data mapping on the mixed traffic grid area based on the historical mixed traffic element dataset to obtain a grid area traffic element dataset; The grid area traffic element dataset is used to perform element evaluation and identification on each grid area in the mixed traffic grid area to obtain the mixed traffic area division element parameter set.

4. The method for dynamic assessment of mixed traffic resilience in a high-speed construction zone according to claim 2, wherein: The obtained M mixed traffic areas include: Performing resilience impact assessment and classification reference sorting on each element information in the traffic area division element set to obtain a traffic area division element reference sequence; Determining a weight factor of a division element according to a reference sequence of the traffic area division elements; Performing regional cluster analysis based on the mixed traffic area division element parameter set to obtain traffic area cluster analysis results; The traffic area cluster analysis result is evaluated and divided into key areas according to the division factor, so as to obtain the M mixed traffic areas.

5. The method for dynamic assessment of mixed traffic resilience in a high-speed construction zone according to claim 1, characterized in that: The obtained M mixed traffic association feature sets include: Mapping the M mixed traffic flow data to the radar and visual processing dual lines for data synchronization parsing to obtain M mixed traffic radar data and M mixed traffic visual data; Tracking vehicle trajectories on the M mixed traffic radar data based on the radar data processing circuit to determine M traffic vehicle operation feature sets; Using the visual data processing circuit to perform vehicle identification and classification on the M mixed traffic visual data to obtain M traffic vehicle behavior feature sets; The M traffic vehicle operation feature sets and the M traffic vehicle behavior feature sets are aligned and fused to obtain the M mixed traffic association feature sets.

6. A dynamic assessment method for mixed traffic resilience in a high-speed construction zone according to claim 5, characterized in that: The determining of M traffic vehicle operation feature sets includes: According to the radar data processing circuit, training and calling the radar target detection network; Using the radar target detection network to perform vehicle target detection and identification code assignment on the M mixed traffic radar data, and output a target coded vehicle set; Initializing a Kalman filter, and performing prediction tracking on the target coded vehicle set based on the Kalman filter to obtain a target vehicle trajectory set; Correlated operation features are extracted based on the target vehicle trajectory set to determine M traffic vehicle operation feature sets, where the M traffic vehicle operation feature sets include traffic flow, traffic speed, and traffic density.

7. The method for dynamic assessment of mixed traffic resilience in a high-speed construction zone according to claim 1, characterized in that: The construction of a dynamic assessment channel for mixed traffic resilience includes: Performing correlation index decomposition and feature data mining on the hybrid traffic resilience assessment system to obtain a hybrid traffic historical resilience index dataset; Using a deep neural network structure to perform resilience assessment training on the mixed traffic history resilience index dataset to obtain a resilience index assessment branch channel; Performing a decision impact analysis on the resilience index evaluation branch channel to determine a resilience index channel decision weight factor; Based on the resilience index channel decision weight factor, the resilience index evaluation branch channels are weighted and combined in parallel to construct the mixed traffic resilience dynamic evaluation channel.

8. A dynamic assessment method for mixed traffic resilience in a high-speed construction zone according to claim 7, characterized in that: The obtaining of the mixed traffic historical resilience indicator dataset includes: Decomposing the hybrid traffic resilience assessment system into related indicators to obtain a set of hybrid traffic resilience related assessment indicators; determining a resilience indicator association depth set according to the resilience indicator channel decision weight factor; Traffic feature data mining is performed based on the resilience index association depth set to obtain the mixed traffic historical resilience index data set.

9. The method for dynamic assessment of mixed traffic resilience in a high-speed construction zone according to claim 1, characterized in that: The determination of the target mixed traffic resilience assessment results includes: Performing a resilience impact range assessment on each traffic area in the M mixed traffic areas to obtain resilience impact coefficients of the M traffic areas; The M mixed traffic resilience assessment information is integrated and analyzed based on the M traffic area resilience influence coefficients to determine the target mixed traffic resilience assessment result.

10. A dynamic assessment system for mixed traffic resilience in high-speed construction zones, characterized by: A method for dynamically evaluating mixed traffic resilience in a high-speed construction zone, for implementing any one of claims 1 to 9, comprising: The regional division and monitoring deployment module is used to divide the mixed traffic area of ​​the target highway construction zone into key areas, obtain M mixed traffic areas, and sequentially deploy integrated radar and visual monitoring equipment in the M mixed traffic areas; A data acquisition module, configured to acquire M mixed traffic flow data through the integrated radar and visual monitoring equipment, and establish a radar and visual processing dual circuit, wherein the radar and visual processing dual circuit includes a radar data processing circuit and a visual data processing circuit; A feature analysis module, configured to perform feature analysis on the M mixed traffic flow data in parallel using the radar visual processing dual lines to obtain M mixed traffic correlation feature sets; An assessment capacity building module is used to build a hybrid transport resilience assessment system, conduct indicator assessment training based on the hybrid transport resilience assessment system, and construct a dynamic assessment channel for hybrid transport resilience; An evaluation and result integration module is used to perform resilience evaluation on the M mixed traffic associated feature sets using the mixed traffic resilience dynamic evaluation channel to obtain M mixed traffic resilience evaluation information, integrate and analyze the M mixed traffic resilience evaluation information, and determine the target mixed traffic resilience evaluation result.

Images