CN120526599A

CN120526599A - Transportation operation monitoring, early warning and decision analysis method based on vehicle-road collaboration

Info

Publication number: CN120526599A
Application number: CN202511021418.5A
Authority: CN
Inventors: 郑晋元
Original assignee: Changan University
Current assignee: Changan University
Priority date: 2025-07-24
Filing date: 2025-07-24
Publication date: 2025-08-22
Anticipated expiration: 2045-07-24
Also published as: CN120526599B

Abstract

The present invention discloses a method for monitoring, warning, and decision-making analysis of transportation operations based on vehicle-road collaboration, which relates to the field of intelligent transportation technology. Through a multi-dimensional perception network of vehicle-side, roadside, and environmental factors and V2X communication, the present invention achieves real-time collection of dynamic traffic elements, ensures the uniformity of spatiotemporal benchmarks for multi-source data, constructs a "road-vehicle-environment-event" semantic network, and combines hierarchical feature extraction to achieve multi-dimensional identification of abnormal events such as accidents, congestion, and severe weather. This improves recognition accuracy, predicts the propagation paths of abnormal events, and possesses the ability to predict global road network trends. Based on a comprehensive risk index, it generates differentiated warnings and triggers multi-level responses such as traffic signal adjustment and route planning, significantly shortening response time, optimizing emergency response efficiency, and establishing a real-time feedback mechanism for decision execution. System performance is continuously optimized as data accumulates, avoiding the drawback of big data platforms lacking an intelligent decision-making closed loop.

Description

Traffic transportation operation monitoring early warning and decision analysis method based on vehicle-road cooperation

Technical Field

The invention relates to the technical field of intelligent transportation, in particular to a traffic transportation operation monitoring, early warning and decision analysis method based on vehicle-road cooperation.

Background

With the rapid development of intelligent traffic systems, conventional traffic operation monitoring and decision-making technologies face a plurality of challenges, for example, chinese patent with bulletin number CN117727183B discloses an automatic driving safety early warning method and system combining with vehicle-road cooperation, and the method comprises: and reading target real-time driving data, then reading a driving association state data set, obtaining a target driving visual model, then carrying out static vehicle-road cooperative safety detection, obtaining a target static cooperative safety detection result, then carrying out dynamic vehicle-road cooperative safety detection, obtaining a dynamic cooperative safety detection result, generating a safety analysis report, and sending the safety analysis report to a safety precaution device for safety precaution. The method solves the problem that the existing method excessively depends on historical data, which increases the difficulty of real-time early warning, possibly reduces early warning accuracy, and cannot be adaptively adjusted according to scene changes, so that the real-time performance of early warning effect is limited. And judging whether the vehicle has safety risk or illegal action in real time according to the dynamic cooperative safety detection result, and carrying out early warning, thereby improving the real-time performance and accuracy of the data.

The patent improves the data real-time performance and early warning accuracy, but still has the following problems:

1. The existing transportation system is limited to the vehicle and the surrounding local environment data, lacks comprehensive perception of road infrastructure, macroscopic road network state and environment parameters, and is difficult to construct a complete traffic running situation map.

2. The risk assessment is only aimed at the safety risk or the illegal behavior of the vehicle individual, does not carry out multidimensional identification on abnormal events such as accidents, congestion, bad weather and the like, and cannot predict the propagation path and the comprehensive risk of the event in the road network.

3. Semantic representation and dynamic modeling of traffic states are not established, influence of environmental factors on a traffic system cannot be quantified, and fine traffic management and decision making are difficult to support.

Disclosure of Invention

The invention aims to provide a traffic transportation operation monitoring early warning and decision analysis method based on vehicle-road cooperation, which utilizes a four-dimensional semantic network and a layered feature extraction technology, combines a graph neural network and time sequence data mining, accurately identifies abnormal events and predicts propagation paths, builds a comprehensive risk assessment model, improves intelligent analysis capability on complex traffic scenes, and realizes efficient linkage of early warning, decision making and execution so as to solve the problems in the background technology.

In order to achieve the above purpose, the present invention provides the following technical solutions:

a traffic transportation operation monitoring, early warning and decision analysis method based on vehicle-road cooperation comprises the following steps:

Constructing a multidimensional sensing network, namely acquiring vehicle state, road environment and infrastructure data in real time, and fusing the acquired vehicle state, road environment and infrastructure data to generate multidimensional space data;

The dynamic traffic state analysis is to conduct layered feature extraction on the merged multidimensional space data, construct a dynamic traffic state model, observe the degree of deviation of traffic flow parameters from normal distribution in real time, and conduct multidimensional identification on abnormal events by combining historical observation data and real-time observation values;

Performing association analysis on the identified abnormal event based on the identification result, predicting a propagation path, and calculating the comprehensive risk index of the abnormal event according to the hazard degree and the propagation risk of the abnormal event;

triggering a multi-level response mechanism based on the comprehensive risk index, generating corresponding early warning information, establishing a decision scheme, performing security verification on the decision scheme, and pushing decision instructions to the vehicle and each infrastructure through a two-way communication link.

Further, the multi-dimensional perception network construction further comprises:

establishing a bidirectional communication link between vehicles and between the vehicles and each infrastructure based on the V2X communication component installed on the vehicles, and establishing connection with the traffic cloud platform based on the V2X communication component;

The vehicle sends vehicle state data packets to adjacent vehicles according to a preset period based on a bidirectional communication link, and each road side device sends event information in real time to form a dynamic communication network between vehicles and between the vehicles and each infrastructure;

determining time synchronization between vehicles and between the vehicles and the infrastructures based on a clock synchronization mechanism;

The vehicle acquires monitoring information of the adjacent vehicle based on the bidirectional communication link, wherein the monitoring information comprises real-time position, motion state and intention information of the adjacent vehicle, and meanwhile, traffic signal state and road environment data actively uploaded by the road side infrastructure are acquired.

Further, the dynamic traffic state analysis further comprises the steps of constructing a four-dimensional semantic network topology structure:

constructing a structural framework of a road network topological graph based on topological attributes and historical traffic characteristics by taking a road physical entity as a topological base node;

Mapping the real-time track of the vehicle into dynamic edges, taking the vehicle position transfer in the adjacent time slices as the connection relation of the edges, and constructing an edge set based on edge attribute behavior characteristics;

The environment parameters are used as weight factors of the graph structure, mapped to corresponding road nodes and vehicle behavior edges according to the time-space grids, and embedded into a four-dimensional semantic network topological structure for the identified abnormal event;

Meanwhile, the association relation among all nodes in the four-dimensional semantic network topology structure is learned, and semantic representation of traffic states is constructed.

Further, the dynamic traffic state analysis hierarchical feature extraction further includes:

Performing space discretization processing on the road network topological graph, dividing the road network into space-time grid units based on the physical entity coordinate range of the road and lane division, extracting the basic characteristics of traffic flow in each space-time grid unit, and forming a space grid traffic state vector;

Extracting edge attributes formed by vehicle position transfer in adjacent time slices, generating a characteristic sequence representing vehicle movement time sequence change, and simultaneously, combining historical traffic data, identifying traffic flow periodicity rules and sudden change characteristics, and generating a time sequence dynamic characteristic vector;

And carrying out weighted fusion on the space grid traffic state vector and the time sequence dynamic feature vector to form a composite feature representation, and embedding the composite feature representation into the four-dimensional semantic network topological structure.

Further, the comprehensive risk prediction evaluation performs association analysis, and specifically includes:

Acquiring historical abnormal event data, extracting propagation modes of different types of abnormal events under different environmental parameters by combining multidimensional space data and environmental parameters in the four-dimensional semantic network topological structure, and constructing a propagation rule knowledge base;

The identified abnormal event is used as a key node, and based on the association relation and the edge attribute among the nodes in the four-dimensional semantic network topological structure, road nodes and adjacent event nodes which are possibly affected are screened out based on a propagation rule knowledge base, and an event evolution tree is constructed;

The tree node attributes comprise event types, space-time positions, current influence degrees and severity degrees, and the tree node connection relation attributes comprise event propagation directions, propagation probability and time delay thresholds.

Further, predicting the propagation path specifically includes:

taking an abnormal event key node as a target node, carrying out multipath sampling on connection relation attributes in an event evolution tree to generate at least one propagation path, wherein each sampling is carried out, and a corresponding branch path is selected according to the propagation probability distribution of the current target node;

Acquiring a time sequence of each target node in each propagation path affected by the same event in combination with the propagation time distribution of the similar event of the history, and generating a propagation path set;

mapping the propagation probability distribution information of each propagation path in the propagation path set to space-time grid units, and performing accumulated calculation on the affected probability of each grid unit in the future period to generate a risk probability cloud picture of the future period;

and identifying key pivot nodes in the propagation path set, calculating risk contribution degrees of the key pivot nodes in the propagation process, and sequencing the priority of each key pivot node based on the risk contribution degrees.

Further, identifying a key hub node in the propagation path set specifically includes:

extracting node attribute and topology structure information of each road node in the propagation path set in a four-dimensional semantic network topology structure;

acquiring basic flow characteristics of each road node and flow conduction sensitivity among the road nodes based on node attributes and topological structure information, and determining a flow weight coefficient and a flow conduction weight coefficient;

Determining a road node law enforcement management weight based on a management level corresponding to a law enforcement management mode according to preset law enforcement management mode information corresponding to each road node;

And comprehensively calculating law enforcement management weight, flow weight coefficient and flow conduction weight coefficient to obtain a road topology weight value, comparing the road topology weight value with a preset weight threshold value, and screening out road nodes higher than the preset weight threshold value as key hub nodes.

Further, the multi-level response mechanism includes:

The method comprises the steps of establishing a mapping rule of an early warning level and a response strategy, wherein the mapping rule comprises the steps that the early warning level is low in risk, text early warning is pushed to surrounding vehicles, the surrounding vehicles are carefully driven through vehicle navigation prompt, the early warning level is medium in risk, road side signal lamps are synchronously triggered to optimize, route suggestions are sent to logistics motorcades, the early warning level is high in risk, cross-regional emergency linkage is started, event details and treatment plans are pushed to a traffic management center, and meanwhile speed limiting instructions are forced through a two-way communication link.

Further, before fusing the acquired vehicle state, road environment and infrastructure data, the method further comprises the steps of reducing noise of image data included in the road environment data;

the noise reduction of the image data in the road environment data comprises the following steps:

an image in road environment data is arbitrarily acquired and used as an image to be noise reduced;

Carrying out gray scale processing on the image to be noise reduced to obtain a gray scale image;

taking one pixel point in the gray level image as a first pixel point, and taking the first pixel point as a center and a preset distance as a radius to determine a target area;

Calculating the difference value between the first pixel point and other pixel points except the first pixel point in the target area to obtain a plurality of difference values;

Traversing all pixel points in the gray level image, and counting the number of abnormal marks of each pixel point;

comparing the number of the abnormal marks with a preset abnormal mark threshold value, and taking the pixel points with the number of the abnormal marks being greater than or equal to the preset abnormal mark threshold value as first abnormal pixel points to obtain a plurality of first abnormal pixel points;

performing edge recognition on the gray level image to determine edge pixel points in the gray level image;

subtracting the first abnormal pixel points from the edge pixel points to obtain second abnormal pixel points;

calculating the dispersion of each second abnormal pixel point respectively, comparing the dispersion with a preset dispersion threshold value, and taking the pixel point when the dispersion is larger than or equal to the preset dispersion threshold value as a third abnormal pixel point to obtain a plurality of third abnormal pixel points;

deleting a plurality of third abnormal pixel points to obtain a noise-reduced gray image;

and traversing all images in the road environment data to obtain the noise-reduced road environment data.

Further, according to the hazard degree and the propagation risk of the abnormal event, calculating the comprehensive risk index of the abnormal event, wherein the comprehensive risk index comprises the following steps of 1-2:

step 1, acquiring an average risk value of a vehicle a in historical driving data and a propagation risk value of an abnormal event on a driving path of the vehicle a, and determining a comprehensive risk index of the abnormal event on the vehicle a based on the average risk value, the propagation risk value and an expected risk value of the abnormal event;

;

wherein, the N represents the total number of risk factor categories in the abnormal event; Representing an average risk value of the vehicle a in the historical driving data; A propagation risk value indicating a traveling path of the vehicle a by an abnormal event; A risk degree value representing an i-th type of risk factors in the n-type of risk factors; a probability density function representing the risk of occurrence of the class i risk factor category; representing the cumulative probability of risk occurrence of the i-th type of risk factor in the next unit time interval from time t; Representing the total cumulative probability of risk of occurrence of the type i risk factor for all times in the future, starting at time t;

Step 2, determining the comprehensive risk index of the abnormal event based on the comprehensive risk index of the abnormal event on the vehicle a and the total number of vehicles in the running path of the vehicle a;

;

wherein, the A composite risk index representing an abnormal event; indicating the total number of vehicles in the travel path along which the vehicle a is located.

Compared with the prior art, the invention has the beneficial effects that:

The real-time acquisition of dynamic traffic elements is realized through a multidimensional sensing network of vehicle ends, road sides and environments and V2X communication, unified space-time basis of multi-source data is ensured, a 'road-vehicle-environment-event' semantic network is constructed, multi-dimensional recognition of abnormal events such as accidents, congestion, bad weather and the like is realized by combining layered feature extraction, recognition accuracy is improved, an abnormal event propagation path can be predicted, global road network situation prediction capability is provided, differential early warning is generated based on comprehensive risk indexes, multi-stage response such as traffic signal adjustment and path planning is triggered, response time is greatly shortened, emergency response efficiency is optimized, a real-time feedback mechanism of decision execution effect is established, system performance is continuously optimized along with data accumulation, and the defect that a large data platform lacks an intelligent decision closed loop is avoided.

Drawings

FIG. 1 is a flow chart of a traffic operation monitoring, early warning and decision analysis method of the invention.

Detailed Description

The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.

In order to solve the problems that the prior art is limited to the vehicle and the surrounding local area, the risk assessment lacks multidimensional recognition and propagation prediction, the decision mechanism lacks differentiated response and cross-regional cooperation, the traffic state semantic representation is not established, and the like, the technical problems that a complete situation map is difficult to construct and the support fine management are difficult to realize, referring to fig. 1, the embodiment provides the following technical scheme:

Constructing a multidimensional sensing network covering a vehicle end, a road side and an environment, acquiring vehicle state, road environment and infrastructure data in real time, fusing the acquired vehicle state, road environment and infrastructure data, generating multidimensional space data containing a time stamp, and realizing the real-time acquisition of dynamic traffic elements and unification of space-time references;

The dynamic traffic state analysis comprises the steps of extracting layered characteristics of the fused multidimensional space data, constructing a dynamic traffic state model, observing the degree of deviation of traffic flow parameters from normal distribution in real time, and carrying out multidimensional identification of accidents, congestion, bad weather and the like on abnormal events by combining historical observation data and real-time observation values;

Triggering a multi-level response mechanism based on a comprehensive risk index, generating early warning information corresponding to the event type, the influence range and the treatment suggestion, constructing a multi-objective optimization model comprising traffic efficiency, safety risk and energy consumption cost, adopting a multi-agent reinforcement learning (MARL) algorithm to establish a differential decision scheme, carrying out safety verification on the decision scheme, establishing a real-time feedback mechanism of a decision execution effect, realizing collaborative updating of a cross-region decision model, improving global adaptability, and pushing decision instructions to vehicles and various infrastructures through a bidirectional communication link;

in this embodiment, the multi-level response mechanism includes:

The method comprises the steps of establishing a mapping rule of an early warning level and a response strategy, wherein the mapping rule comprises the steps that the early warning level is low in risk, text early warning (such as 'front road section wet skid') is pushed to surrounding vehicles, the surrounding vehicles are carefully driven through vehicle navigation prompt, the early warning level is medium in risk, road side signal lamp timing optimization is synchronously triggered (such as green light extension and congestion relief) and route advice is sent to logistics motorcades, the early warning level is high in risk, cross-region emergency linkage is started, event details and treatment plans are pushed to traffic management centers, and meanwhile speed limiting instructions (such as 40km/h of accident road section speed limit) are forced through a two-way communication link.

In the embodiment, the multi-dimensional perception network construction realizes multi-source data fusion and space-time unification, a vehicle-road-environment multi-dimensional fusion perception system is constructed, the dynamic traffic state analysis can identify abnormal events in a multi-dimensional manner, the layered characteristic extraction and the multi-dimensional anomaly identification model are used for realizing parallel detection of accidents, congestion and bad weather, the comprehensive risk prediction evaluation can accurately predict propagation paths and risks, the problem of one-sided performance of the traditional risk evaluation is solved, the influence range of the predictable events is expanded, the multi-stage response mechanism in early warning generation and decision can be adopted, differential measures can be adopted according to the risk level, the multi-objective optimization model and MARL algorithm are applied, the decision is more scientific, and the global adaptability is improved through regional collaborative updating.

In this embodiment, the multidimensional aware network construction further includes:

the vehicle sends vehicle state data packets to adjacent vehicles according to a preset period based on a bidirectional communication link, wherein the vehicle state data packets comprise beacon frames of positions, speeds, course angles and the like, and each road side device sends event information such as traffic signal states, road construction/congestion and the like in real time to form dynamic communication networks among vehicles and between the vehicles and each infrastructure;

determining time synchronization between vehicles and between the vehicles and each infrastructure based on an IEEE1588 clock synchronization protocol or a GPS Beidou satellite clock synchronization mechanism, and realizing nanosecond time synchronization of all equipment;

In this embodiment, the dynamic traffic state analysis further includes constructing a four-dimensional semantic network topology of "road-vehicle-environment-event":

Constructing a structural skeleton of a road network topological graph based on topological attributes (coordinate range, lane number and speed limit value) and historical traffic characteristics (daily traffic and traffic speed distribution) by taking road physical entities (such as road sections and intersections) as topological foundation nodes;

Mapping the real-time track of the vehicle into a dynamic edge, taking the vehicle position transition in the adjacent time slices as the connection relation of the edge, and constructing an edge set reflecting the traffic flow dynamic interaction based on the behavior characteristics of the edge attribute including speed, acceleration, steering intention and the like;

Taking environmental parameters (precipitation intensity, visibility and road surface temperature) as weight factors of a graph structure, mapping the weight factors to corresponding road nodes and vehicle behavior edges according to a time-space grid, realizing quantitative influence characterization of the environmental factors on traffic states, and embedding identified abnormal events into a four-dimensional semantic network topological structure, wherein node attributes comprise event types, influence ranges and timeliness parameters;

Meanwhile, the association relation among all nodes in the four-dimensional semantic network topological structure is learned through a Graph Neural Network (GNN), so that semantic representation of the traffic state is constructed.

In the embodiment, a dynamic communication network is constructed based on the V2X communication component, nanosecond time synchronization is realized by combining IEEE1588 or satellite clock synchronization, the difficulty of traditional communication delay and time deviation is overcome, the real-time performance and the synergy of data are remarkably enhanced, the traffic state is visually represented by a four-dimensional semantic network topological structure of road-vehicle-environment-event, the dynamic interaction of traffic flow and the influence of environmental factors can be more accurately represented, the abnormal event identification is more comprehensive, the understanding accuracy of a complex traffic scene is improved, the prediction error of an event propagation path is reduced, and more visual and accurate semantic representation is provided for risk assessment and decision.

In this embodiment, the dynamic traffic state analysis hierarchical feature extraction further includes:

performing space discretization processing on the road network topological graph, dividing the road network into space-time grid units containing geographic coordinates, speed limit values, lane numbers and other attributes based on the physical entity coordinate range of the road and lane division, extracting traffic flow basic features (flow, occupancy rate and average speed) in each space-time grid unit, and forming a space grid traffic state vector;

And carrying out weighted fusion on the space grid traffic state vector and the time sequence dynamic feature vector to form a composite feature representation containing space-time context, highlighting feature influence (such as real-time influence weight of an accident road section on a peripheral road network) of a key space-time area, and embedding the four-dimensional semantic network topology structure.

In the embodiment, the space-time grid unit division and feature fusion strategy is used for performing discretization on road network space, combining vehicle motion time sequence features and historical data rules, constructing composite feature representation comprising space-time context, accurately capturing the influence of a key region, embedding a four-dimensional semantic network topological structure, realizing deep fusion analysis of space distribution and time dynamic change, refining the extraction granularity of traffic flow basic features to a lane level, and combining the historical data recognition rules, so that the sudden change of traffic flow can be predicted greatly in advance, and the composite feature representation effectively improves the influence weight accuracy of the key region such as an accident road section and the like, thereby improving the timeliness and accuracy of traffic state analysis.

In this embodiment, the comprehensive risk prediction evaluation performs association analysis, which specifically includes:

Acquiring historical abnormal event data (such as a diffusion mode of 1000 accidents in the past), extracting propagation modes of different types of abnormal events (accidents, congestion and bad weather) under different environmental parameters (such as rainy days and nights) by adopting a time sequence data mining algorithm in combination with multidimensional space data and environmental parameters in the four-dimensional semantic network topological structure, and constructing a propagation rule knowledge base, wherein the content of the rule knowledge base comprises the mapping relation of event types, environmental conditions and propagation paths;

The tree node attributes comprise event type, space-time position, current influence degree and severity, and the tree node connection relation attributes comprise event propagation direction, propagation probability and time delay threshold (for example, the probability of congestion propagating from a road section A to a road section B is 0.7, and the delay time is 5 minutes);

In the embodiment, the initial propagation probability of each side is calculated through an association rule mining algorithm, the average propagation time counted by combining historical data is combined, the time delay in the side attribute is determined, meanwhile, the current road network traffic state parameters (such as actual flow and vehicle speed distribution) and the traffic flow basic characteristics in the space-time grid unit are taken as constraint conditions, branches of an evolution tree are pruned, and a low-probability propagation path is eliminated.

In this embodiment, predicting the propagation path specifically includes:

Acquiring the time sequence of each target node in each propagation path by a time sequence interpolation algorithm in combination with the propagation time distribution of the similar historical events, and generating a propagation path set containing space-time dimensions;

mapping the propagation probability distribution information of each propagation path in the propagation path set to space-time grid units, performing accumulated calculation on the affected probability of each grid unit in a future period, generating a risk probability cloud chart of the future period, and intuitively displaying the affected probabilities of different areas at different time points;

And identifying key hub nodes in the propagation path set, calculating the risk contribution degree of the key hub nodes in the propagation process (such as 40% of road network traffic capacity is reduced due to the fact that the key hub nodes are blocked), and sequencing the priority of each key hub node based on the risk contribution degree to serve as a priority treatment object of subsequent response.

In this embodiment, identifying key hub nodes in the propagation path set specifically includes:

In the embodiment, the traffic flow corresponding to a key time node is obtained from the road node attribute, wherein the key time node comprises abnormal event occurrence time and a high occurrence period of a historical similar event;

in this embodiment, topology structure information is extracted, including the number of road nodes having an association relationship with the road node, and percentage data of traffic volume change of the associated road node when the traffic volume of the road node increases by 1%, where the data is obtained by analyzing historical propagation data of edge attributes in an event evolution tree;

In the embodiment, taking the average value of traffic flow of each road node at a key time node as a basic flow characteristic, calculating the average value of the variation percentages of the associated road nodes corresponding to the traffic flow variation of the road node based on the variation percentages of the traffic flow of each associated road node, and quantifying the flow transmission sensitivity among the road nodes;

In this embodiment, the ratio calculation is performed between the average traffic flow value of each road node and a preset traffic flow reference value (such as the average traffic flow of the road network or the design traffic of the road section), so as to obtain a flow weight coefficient, and reflect the relative importance of the traffic flow of the road node;

in the embodiment, the average value of the change percentages of the associated road nodes of each road node is normalized to obtain a flow conduction weight coefficient, and the influence diffusion capacity of the road nodes on the peripheral road network is represented;

Determining the law enforcement management weight of the road nodes according to the preset law enforcement management mode information (such as daily law enforcement frequency and emergency disposal priority) corresponding to each road node and based on the management level corresponding to the law enforcement management mode through a mapping function, wherein the road nodes with high law enforcement priority are given higher weight, and the association of management resource investment and importance of the road nodes is reflected;

In the embodiment, based on a four-dimensional semantic network, a time sequence mining algorithm is used for establishing an abnormal event propagation rule knowledge base by combining historical data and real-time parameters, so that the abnormal event propagation mode identification accuracy is effectively improved, the dynamic generation and accurate prediction of propagation paths are realized by an event evolution tree and a multi-path sampling and pruning algorithm, the prediction efficiency is greatly improved, the risk space-time distribution can be intuitively displayed, visual decision support is provided for traffic management, key hub nodes are evaluated based on multi-dimensional weight coefficients, the risk contribution degree of road nodes is quantized and ordered, basis is provided for accurate early warning and priority disposal, the emergency resource allocation efficiency is improved, the traffic capacity guarantee rate of key areas is improved, the integral influence of abnormal events on a traffic system is effectively reduced, and the traffic risk prevention and control capacity is remarkably enhanced.

In the embodiment, before fusing the acquired vehicle state, road environment and infrastructure data, the method further comprises the steps of reducing noise of image data included in the road environment data;

In this embodiment, the pixels with difference exceeding the threshold are marked as abnormal by calculating the gray difference between the central pixel and other pixels in a certain range around, which is essentially to capture the pixels with significantly abrupt change of the gray value in the local area, and distinguishing from the pixels with directly determining abnormal pixel points in the prior art means that final judgment is made on the pixel properties (normal/abnormal), if the threshold is not reasonably set or the area range is improperly selected later, the pixels which have been determined as abnormal permanently lose their original properties (such as gray value, spatial position, etc.), and are difficult to retrospectively adjust, while the pixels are only marked with a temporary label (such as "to-be-verified abnormal"), which retains all information of the original pixels, and can be re-evaluated later by adjusting the threshold, expanding/shrinking the target area, etc., so as to avoid irreversible errors caused by the initial parameter errors.

In this embodiment, the gray-scale image is edge-identified, in particular by a pre-trained edge-identification model.

In the embodiment, the subtraction is performed on the plurality of first abnormal pixel points and the edge pixel points to obtain a plurality of second abnormal pixel points, specifically, deleting the pixel points which are repeated with the edge pixel points in the plurality of first abnormal pixel points, and taking the rest pixel points as the second abnormal pixel points.

In the embodiment, the specific implementation mode of calculating the dispersion of each second abnormal pixel point is that the minimum Euclidean distance between each second abnormal pixel point and the edge pixel point is used as the dispersion of each second abnormal pixel point, and the average value of the dispersion of all second abnormal pixel points is used as a preset dispersion threshold value.

The technical scheme has the beneficial effects that an image is selected from road environment data at will to be used as an image to be noise reduced, and then gray processing is carried out on the image to convert a color image into a gray image. The purpose of the method is to simplify the image data, and facilitate the subsequent pixel analysis, because the gray image only contains brightness information, the complexity of the data is reduced; for each pixel point in the gray image (one is taken as a first pixel point firstly), a target area is determined by taking the pixel point as a center and taking a preset distance as a radius, the difference value between the first pixel point and other pixel points in the target area is calculated, abnormal pixel points are marked by comparing the difference value with a preset difference value threshold, the process is to preliminarily judge which pixel points are possible noise points based on the brightness difference of the pixel points in the local area, because the brightness difference value of adjacent pixel points is not excessive under normal conditions, all the pixel points in the whole gray image are traversed, and the abnormal marking quantity of each pixel point is counted. The method comprises the steps of comparing the number of abnormal marks with a preset abnormal mark threshold value to find out a first abnormal pixel point, screening out pixel points which are possibly noise, if one pixel point is marked as abnormal in a plurality of local areas, determining the third abnormal pixel point and deleting the pixel points, carrying out edge identification on a gray image to determine the edge pixel point, carrying out subtraction operation on the first abnormal pixel point and the edge pixel point to obtain a second abnormal pixel point, wherein the second abnormal pixel point is possibly misjudged as a noise point due to large brightness change of the edge pixel point, eliminating interference of the edge pixel point in the mode, respectively calculating the dispersion of each second abnormal pixel point, reflecting the distribution characteristics of the pixel points in the local areas, comparing the dispersion with the preset dispersion threshold value, determining the third abnormal pixel point and deleting the pixel points, thus obtaining a noise-reduced gray image, finally traversing all images in road environment data, completing noise reduction of the image part, accurately identifying and removing noise points through detailed pixel point analysis and screening processes, and removing noise points in the image by a plurality of special noise points by taking the noise-reducing method into consideration, and the noise-reducing factors in the local areas of the image is better than the local areas.

In this embodiment, according to the hazard degree and the propagation risk of the abnormal event, the comprehensive risk index of the abnormal event is calculated, including the steps of 1-2:

;

In this embodiment, the propagation probability distribution information of each propagation path in the propagation path set may determine a propagation risk value of the travel path of the vehicle a.

The technical scheme has the advantages that factors in multiple aspects including historical risk values of the vehicles, propagation risks of abnormal events on a driving path, risk degrees of various risk factors, probability of occurrence of risks of various risk factors and the like are comprehensively considered, the multi-dimensional evaluation can reflect risk conditions of the abnormal events more comprehensively, one-sided performance of single factor evaluation is avoided, time factors are introduced through time integration in a probability density function, probability of occurrence of risks of the risk factors can be evaluated according to different time intervals, accordingly, conditions of time variation of risks are well adapted, comprehensive risk indexes of single vehicles are calculated, comprehensive risk indexes (T) of the abnormal events are obtained through comprehensive calculation of all the vehicles, risks of the single vehicles in the abnormal events can be evaluated, risk influence of the abnormal events on a vehicle group on the whole driving path can be grasped, and the method is beneficial to formulating a targeted risk management strategy.

The foregoing is only a preferred embodiment of the present invention, but the scope of the present invention is not limited thereto, and any person skilled in the art, who is within the scope of the present invention, should be covered by the protection scope of the present invention by making equivalents and modifications to the technical solution and the inventive concept thereof.

Claims

1. A transportation operation monitoring, early warning, and decision-making analysis method based on vehicle-road collaboration, characterized by including:

Multi-dimensional perception network construction: Build a multi-dimensional perception network to obtain real-time vehicle status, road environment and infrastructure data, and integrate the collected vehicle status, road environment and infrastructure data to generate multi-dimensional spatial data;

Before fusing the collected vehicle status, road environment, and infrastructure data, the process also includes denoising the image data included in the road environment data, traversing all images in the road environment data, and obtaining the denoised road environment data.

Dynamic traffic status analysis: Perform hierarchical feature extraction on the fused multi-dimensional spatial data to construct a dynamic traffic status model. Observe in real time the degree to which traffic flow parameters deviate from normal distribution. Combine historical observation data with real-time observations to perform multi-dimensional identification of abnormal events.

Comprehensive risk prediction and assessment: Based on the identification results, correlation analysis is performed on the identified abnormal events, the propagation path is predicted, and the comprehensive risk index of the abnormal events is calculated according to the degree of harm and propagation risk of the abnormal events;

Warning generation and decision-making: Based on the comprehensive risk index, a multi-level response mechanism is triggered, and corresponding warning information is generated. A decision plan is established and the safety of the decision plan is verified. The decision instructions are pushed to vehicles and various infrastructure through a two-way communication link.

2. The method for monitoring, early warning, and decision-making analysis of transportation operations based on vehicle-road collaboration according to claim 1, wherein the multi-dimensional perception network construction further comprises:

Building bidirectional communication links between vehicles and between vehicles and infrastructure based on the V2X communication components installed in the vehicles, and establishing a connection with the traffic cloud platform based on the V2X communication components;

The vehicle sends vehicle status data packets to adjacent vehicles based on a two-way communication link at a preset period, and each roadside device sends event information in real time, forming a dynamic communication network between vehicles and between vehicles and various infrastructures;

Determine time synchronization between vehicles and between vehicles and infrastructure based on clock synchronization mechanism;

The vehicle obtains monitoring information of adjacent vehicles based on a two-way communication link, including the real-time position, movement status and intention information of adjacent vehicles, and at the same time collects traffic signal status and road environment data actively uploaded by roadside infrastructure.

3. The method for monitoring, early warning, and decision-making analysis of transportation operations based on vehicle-road collaboration according to claim 2, wherein the dynamic traffic status analysis further comprises constructing a four-dimensional semantic network topology structure:

Taking the physical entities of roads as the topological basic nodes, the structural skeleton of the road network topology map is constructed based on topological attributes and historical traffic characteristics;

The real-time trajectory of the vehicle is mapped into a dynamic edge, and the vehicle position transfer within adjacent time slices is used as the edge connection relationship. Based on the edge attribute behavior characteristics, an edge set is constructed;

Environmental parameters are used as weight factors in the graph structure, mapped to corresponding road nodes and vehicle behavior edges according to the spatiotemporal grid, and the identified abnormal events are embedded in the four-dimensional semantic network topology structure;

At the same time, the association relationship between nodes in the four-dimensional semantic network topology structure is learned to construct a semantic representation of the traffic status.

4. The method for monitoring, early warning, and decision-making analysis of transportation operations based on vehicle-road collaboration according to claim 3, wherein the dynamic traffic state analysis and hierarchical feature extraction further comprises:

The road network topology is spatially discretized. Based on the physical coordinate range of the road and the lane division, the road network is divided into spatiotemporal grid units. The basic characteristics of traffic flow in each spatiotemporal grid unit are extracted to form a spatial grid traffic state vector.

Extract edge attributes formed by vehicle position transfer within adjacent time slices to generate a feature sequence that represents the temporal changes in vehicle movement. At the same time, combine historical traffic data to identify the periodic patterns and sudden change characteristics of traffic flow and generate a temporal dynamic feature vector.

The spatial grid traffic state vector and the temporal dynamic feature vector are weightedly fused to form a composite feature representation, and then embedded into the four-dimensional semantic network topology structure.

5. The method for monitoring, early warning, and decision-making analysis of transportation operations based on vehicle-road collaboration according to claim 4 is characterized in that the comprehensive risk prediction and assessment is used to perform correlation analysis, specifically including:

Acquire historical abnormal event data, combine it with the multidimensional spatial data and environmental parameters in the four-dimensional semantic network topology structure, extract the propagation patterns of different types of abnormal events under different environmental parameters, and build a propagation rule knowledge base;

Taking the identified abnormal events as key nodes, based on the association relationship and edge attributes between nodes in the four-dimensional semantic network topology structure, and based on the propagation rule knowledge base, the potentially affected road nodes and adjacent event nodes are screened out to construct an event evolution tree;

The attributes of tree nodes include event type, spatiotemporal location, current impact level, and severity level; the attributes of tree node connection relationships include event propagation direction, propagation probability, and time delay threshold.

6. The method for monitoring, early warning, and decision-making analysis of transportation operations based on vehicle-road collaboration according to claim 5, wherein predicting the propagation path specifically includes:

Taking the key node of the abnormal event as the target node, multi-path sampling is performed on the connection relationship attributes in the event evolution tree to generate at least one propagation path. During each sampling, the corresponding branch path is selected according to the propagation probability distribution of the current target node;

Combined with the propagation time distribution of similar historical events, the time series of each target node affected in each propagation path is obtained to generate a propagation path set;

Map the propagation probability distribution information of each propagation path in the propagation path set to the spatiotemporal grid unit, and cumulatively calculate the probability of each grid unit being affected in the future period to generate a risk probability cloud map for the future period;

Identify the key hub nodes in the propagation path set, calculate the risk contribution of the key hub nodes in the propagation process, and prioritize each key hub node based on the risk contribution.

7. The method for monitoring, early warning, and decision-making analysis of transportation operations based on vehicle-road collaboration according to claim 6, wherein identifying key hub nodes in the set of transmission paths specifically includes:

Extract node attributes and topological structure information of each road node in the propagation path set in the four-dimensional semantic network topological structure;

Based on the node attributes and topological structure information, the basic flow characteristics of each road node and the flow transmission sensitivity between road nodes are obtained to determine the flow weight coefficient and flow transmission weight coefficient;

According to the preset law enforcement management method information corresponding to each road node, the law enforcement management weight of the road node is determined based on the management level corresponding to the law enforcement management method;

The law enforcement management weight, flow weight coefficient and flow conduction weight coefficient are comprehensively calculated to obtain the road topology weight value, which is compared with the preset weight threshold to screen out road nodes with a value higher than the preset weight threshold as key hub nodes.

8. The method for monitoring, early warning, and decision-making analysis of transportation operations based on vehicle-road collaboration according to claim 7 is characterized in that the multi-level response mechanism includes:

Establish mapping rules between warning levels and response strategies. For example, if the warning level is low risk, text warnings will be pushed to surrounding vehicles, and cautious driving prompts will be given through the in-vehicle navigation. If the warning level is medium risk, the timing optimization of roadside traffic lights will be triggered simultaneously, and route suggestions will be sent to the logistics fleet. If the warning level is high risk, cross-regional emergency linkage will be initiated, and event details and disposal plans will be pushed to the traffic management center. At the same time, speed limit instructions will be enforced through a two-way communication link.

9. The method for monitoring, early warning, and decision-making analysis of transportation operations based on vehicle-road collaboration according to claim 1, wherein the step of reducing noise of image data in road environment data comprises:

Arbitrarily obtain an image from the road environment data as the image to be denoised;

Performing grayscale processing on the image to be denoised to obtain a grayscale image;

Randomly select a pixel point in the grayscale image as the first pixel point; determine the target area with the first pixel point as the center and a preset distance as the radius;

Calculating the difference between the first pixel and the other pixels in the target area except the first pixel to obtain a plurality of difference values; comparing the plurality of difference values with a preset difference threshold, and marking the pixels whose difference values are greater than or equal to the preset difference threshold as abnormal;

Traverse all pixels in the grayscale image and count the number of abnormal marks for each pixel;

Comparing the number of abnormal marks with a preset abnormal mark threshold, taking pixels whose number of abnormal marks is greater than or equal to the preset abnormal mark threshold as first abnormal pixel points, and obtaining a plurality of first abnormal pixel points;

Perform edge recognition on the grayscale image to determine the edge pixels in the grayscale image;

Subtracting a plurality of first abnormal pixel points from the edge pixel points to obtain a plurality of second abnormal pixel points;

Calculating the discreteness of each second abnormal pixel point respectively, and comparing the discreteness with a preset discreteness threshold, taking the pixel point when the discreteness is greater than or equal to the preset discreteness threshold as a third abnormal pixel point, to obtain a plurality of third abnormal pixel points;

Deleting several third abnormal pixels to obtain a denoised grayscale image;

Traverse all images in the road environment data to obtain the denoised road environment data.

10. The method for monitoring, early warning, and decision-making analysis of transportation operations based on vehicle-road collaboration according to claim 6 is characterized in that a comprehensive risk index of an abnormal event is calculated based on the degree of harm and the risk of propagation of the abnormal event, comprising steps 1-2:

Step 1: Obtain the average risk value of vehicle a in historical driving data and the propagation risk value of abnormal events on the driving path of vehicle a. Based on the average risk value, the propagation risk value, and the expected risk value of the abnormal event, determine the comprehensive risk index of the abnormal event for vehicle a.

;

in, represents the comprehensive risk index of the abnormal event to vehicle a; n represents the total number of risk factors in the abnormal event; Represents the average risk value of vehicle a in historical driving data; represents the risk value of the abnormal event spreading to the driving path of the vehicle a; Represents the risk level value of the i-th risk factor type among n risk factors; The probability density function of the risk of occurrence of the i-th risk factor type; It represents the cumulative probability of the occurrence of risk of the i-th risk factor in the next unit time interval starting from time t; It represents the total cumulative probability of the occurrence of risk of the i-th risk factor in all future time starting from time t;

Step 2: Determine the comprehensive risk index of the abnormal event based on the comprehensive risk index of the abnormal event to vehicle a and the total number of vehicles in the driving path of vehicle a;

;

in, Indicates the comprehensive risk index of abnormal events; Indicates the total number of vehicles in the driving path of vehicle a.