A Multi-Indicator Attribution Analysis Method and System for Traffic Anomalies Based on Intra-Group and Inter-Group Distance

Abstract

This invention belongs to the field of traffic control. To address the problem that existing technologies cannot provide practical suggestions for traffic management and emergency command, it provides a method and system for multi-indicator attribution analysis of traffic anomalies based on intra- and inter-group distances. The method includes calculating a comprehensive safety index based on quantified traffic safety indicators and their corresponding weight values. When abnormal fluctuations in the comprehensive safety index are detected in the current period, the traffic safety index of the current period is compared with that of the previous normal period. By calculating the differences between groups and the dispersion of each traffic safety indicator within groups, the contribution of each traffic safety indicator to the change in the comprehensive safety index is quantified, and the dominant factors are accurately identified. Finally, an interpretable attribution report is generated, which can quantitatively and interpretably perform multi-indicator attribution analysis of traffic anomalies, providing guiding suggestions for traffic management and emergency command.

Description

Technical Field

[0001] This invention belongs to the field of traffic control, and in particular relates to a method and system for multi-indicator attribution analysis of traffic anomalies based on intra-group and inter-group distances. Background Technology

[0002] The statements in this section are merely background information related to the present invention and do not necessarily constitute prior art.

[0003] By utilizing IoT sensors, cameras, and other devices, multi-dimensional traffic flow data (such as average vehicle speed, traffic volume, time occupancy, vehicle spacing, and lane change frequency) can be continuously collected for road segments. The core challenge in achieving precise traffic control and emergency command lies in quickly and accurately identifying the key factors causing sudden situations such as traffic accidents, abnormal congestion, and severe weather from massive amounts of monitoring data.

[0004] Currently, methods for analyzing traffic anomalies include threshold alerts, statistical analysis, and machine learning models. Threshold alerts set a fixed threshold for a single indicator (e.g., vehicle speed below 30 km / h). This method is simple but extremely crude, unable to handle complex situations involving multiple coupled indicators, and cannot quantify the contribution of each indicator, resulting in a high false alarm rate and failing to answer the question of "which indicator is the main cause." Statistical analysis (e.g., correlation analysis) identifies anomalous indicators by calculating the deviation of each indicator from the normal range during abnormal periods (e.g., Z-score), but it only reflects the magnitude of change and cannot reveal causal relationships. A large change in one indicator may only be an "effect" rather than a "cause," and it cannot effectively rank and quantify the contributions of multiple indicators. Machine learning models (e.g., classification models) determine traffic conditions by training models (e.g., SVM, random forest). These models, however, are "black boxes" or "grey boxes," lacking interpretability. Even if an "anomaly" can be identified, it is difficult to clearly and quantitatively explain which characteristics (indicators) and their respective importance led to the anomaly. In other words, it lacks the ability to "attribute" causes, making it difficult for traffic management personnel to understand and trust it, and thus unable to guide specific control measures.

[0005] In summary, existing technologies lack an effective means to quantitatively and interpretably perform multi-indicator attribution analysis of traffic anomalies, and therefore cannot provide practical suggestions for traffic control and emergency command. Summary of the Invention

[0006] To address the aforementioned technical problems, this invention provides a method and system for multi-indicator attribution analysis of traffic anomalies based on intra-group and inter-group distances. This method can quantitatively and interpretably perform multi-indicator attribution analysis of traffic anomalies, providing guidance for traffic control and emergency command.

[0007] To achieve the above objectives, the present invention adopts the following technical solution:

[0008] The first aspect of the present invention provides a multi-indicator attribution analysis method for traffic anomalies based on intra-group and inter-group distances.

[0009] In one or more embodiments, a multi-indicator attribution analysis method for traffic anomalies based on intra-group and inter-group distance is provided, including:

[0010] Based on the point cloud data of the set road segment, vehicle trajectories across frames are established to obtain continuous motion information of each vehicle, and then multi-dimensional monitoring indicators are calculated in real time. The multi-dimensional monitoring indicators include average vehicle speed, traffic flow, time occupancy rate, average headway, and speed difference between lanes.

[0011] Based on multidimensional monitoring indicators, traffic safety indicators are constructed and quantified. Based on the quantified traffic safety indicators and their corresponding weight values, a comprehensive safety index is calculated. The traffic safety indicators include following risk, lane-changing risk, and speed risk.

[0012] When abnormal fluctuations in the overall safety index are detected in the current period, the traffic safety index of the current period is compared with that of the previous normal period. By calculating the differences between groups and the degree of dispersion of each traffic safety index within groups, the contribution of each traffic safety index to the change in the overall safety index is quantified, thereby identifying the dominant factors and generating an interpretable attribution report.

[0013] As one implementation method, the process of quantifying the contribution of each traffic safety indicator to the change in the overall safety index is as follows:

[0014] For each traffic safety indicator, its attribution score is calculated based on the ratio of inter-group variance to intra-group dispersion.

[0015] The attribution scores of each traffic safety indicator are normalized to obtain their contribution and then ranked accordingly.

[0016] As one implementation method, the contribution of each traffic safety indicator to the change in the overall safety indicator. The formula is:

[0017] ;

[0018] in, , are the mean and variance of the i-th traffic safety indicator in the abnormal group, respectively; , denoted as the mean and variance of the i-th traffic safety indicator in the normal group.

[0019] As one implementation method, the corresponding weight of each traffic safety indicator is determined by calculating the information entropy of each traffic safety indicator.

[0020] As one implementation method, following risk is characterized by headway (THW) and time to collision (TTC).

[0021] THW=D / V_ego;

[0022] When V_ego > V_front, TTC = D / (V_ego - V_front) holds true.

[0023] Where D is the distance between the vehicle and the vehicle in front, V_ego is the speed of the vehicle, and V_front is the speed of the vehicle in front.

[0024] As one implementation method, the formula for calculating lane change risk is:

[0025] Lane change risk = α × f(TTC_front) + β × f(TTC_rear);

[0026] When the current vehicle speed is greater than the vehicle in front, TTC_front = d_front / (v_ego - v_front);

[0027] When the speed of the following vehicle is greater than that of this vehicle, TTC_rear = d_rear / (v_rear - v_ego);

[0028] f(TTC) = 1 / TTC;

[0029] Where TTC is the collision time; TTC_front and TTC_rear are the collision times with the vehicles in front and behind in the target lane, respectively; f() is the reciprocal function of the collision time; d_front and d_rear are the distances to the vehicles in front and behind in the target lane, respectively; V_ego is the vehicle speed, and V_front is the speed of the vehicle in front; α and β are the weight values ​​of the corresponding reciprocal functions of the collision time; α+β =1.

[0030] As one implementation method, speed risk is assessed using segment-level speed volatility. To represent it, its expression is:

[0031] ;

[0032] ;

[0033] Where M is the total number of valid vehicles passing through the road segment within the time period; This indicates the average acceleration and deceleration intensity of the vehicle; a higher value indicates more unstable driving behavior. Let be the instantaneous speed of the vehicle at time t; Let N be the instantaneous speed of the vehicle at time t-1; N is the number of valid sampling points of the vehicle within window T.

[0034] A second aspect of the present invention provides a multi-indicator attribution analysis system for traffic anomalies based on intra-group and inter-group distances.

[0035] In one or more embodiments, a multi-indicator attribution analysis system for traffic anomalies based on intra-group and inter-group distance includes:

[0036] The multi-dimensional monitoring index calculation module is used to establish cross-frame vehicle trajectories based on point cloud data of a set road segment, obtain continuous motion information of each vehicle, and then calculate multi-dimensional monitoring indicators in real time; the multi-dimensional monitoring indicators include average vehicle speed, traffic flow, time occupancy rate, average headway, and speed difference between lanes.

[0037] The comprehensive safety index calculation module is used to construct and quantify traffic safety indicators based on multi-dimensional monitoring indicators, and to calculate the comprehensive safety index based on the quantified traffic safety indicators and their corresponding weight values; the traffic safety indicators include following risk, lane-changing risk and speed risk;

[0038] The indicator attribution analysis module is used to compare the traffic safety indicators of the current period with those of the previous normal period when abnormal fluctuations in the comprehensive safety indicators are detected in the current period. By calculating the differences between groups and the degree of dispersion of each traffic safety indicator within the group, the contribution of each traffic safety indicator to the change of the comprehensive safety indicators is quantified, thereby identifying the dominant factors and generating an interpretable attribution report.

[0039] A third aspect of the present invention provides a computer-readable storage medium.

[0040] A computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps in the above-described method for multi-indicator attribution analysis of traffic anomalies based on intra-group and inter-group distances.

[0041] A fourth aspect of the present invention provides an electronic device.

[0042] An electronic device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the steps in the multi-indicator attribution analysis method for traffic anomalies based on intra-group and inter-group distances as described above.

[0043] Compared with the prior art, the beneficial effects of the present invention are:

[0044] This invention constructs and quantifies traffic safety indicators based on multidimensional monitoring indicators, and then calculates a comprehensive safety index by combining their corresponding weight values. When abnormal fluctuations occur in the comprehensive safety index within the current period, the traffic safety index of the current period is compared with that of the previous normal period. By calculating the differences between groups and the dispersion of each traffic safety indicator within a group, the contribution of each traffic safety indicator to the change in the comprehensive safety index is quantified, thereby identifying the dominant factors and generating an interpretable attribution report. This invention uses normal data from the previous period as a baseline, effectively eliminating the influence of external factors such as season, week, and weather, ensuring that the attribution results focus on the real abnormal fluctuations. Moreover, it introduces the dispersion within a group as the denominator, automatically penalizing indicators that are inherently volatile and unstable, avoiding misjudgments, and making the attribution results more robust and reliable. Attached Figure Description

[0045] The accompanying drawings, which form part of this invention, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an improper limitation of the invention.

[0046] Figure 1 This is a flowchart of the multi-indicator attribution analysis method for traffic anomalies based on intra-group and inter-group distance according to an embodiment of the present invention;

[0047] Figure 2 This is a schematic diagram of the structure of a traffic anomaly multi-index attribution analysis system based on intra-group and inter-group distance according to an embodiment of the present invention;

[0048] Figure 3 This is a schematic diagram of an electronic device according to an embodiment of the present invention. Detailed Implementation

[0049] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0050] It should be noted that the following detailed description is illustrative and intended to provide further explanation of the invention. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0051] It should be noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the scope of exemplary embodiments according to the invention. As used herein, the singular form is intended to include the plural form as well, unless the context clearly indicates otherwise. Furthermore, it should be understood that when the terms "comprising" and / or "including" are used in this specification, they indicate the presence of features, steps, operations, devices, components, and / or combinations thereof.

[0052] Terminology Explanation:

[0053] An unusual event refers to an abnormal event that occurs suddenly and deviates from the normal state of traffic operation on a highway. Its manifestations include, but are not limited to: severe congestion, traffic accidents, vehicle breakdowns, adverse weather conditions (such as fog, heavy rain, snow), cargo spillage, and other situations that cause a rapid deterioration of traffic flow.

[0054] Multi-dimensional monitoring metrics refer to multi-dimensional data collected by sensor networks deployed along highways to characterize traffic flow. Key metrics include:

[0055] Average speed: The average speed of all vehicles passing through a road segment over a specified time period.

[0056] Traffic volume: The total number of vehicles passing through a road segment within a specified time period.

[0057] Time occupancy: The percentage of time a sensor is occupied by a vehicle within a specified time period; it is a direct indicator of traffic density.

[0058] Average Time Gap / Distance Gap: The average time or space interval between adjacent vehicles.

[0059] Speed ​​difference between lanes: The difference in average vehicle speed between different lanes, used to assess traffic flow stability and conflict risk.

[0060] Attribution Analysis: In this invention, it is a calculation process used to quantify the contribution of each monitoring indicator to a specific emergency and to rank their importance. Its purpose is to identify the dominant and secondary factors that led to the occurrence of the emergency.

[0061] Within-group / Between-group distance: This is a statistical term and its applied definition in this field.

[0062] Between-group variability: This refers to the degree of difference between the mean of a monitoring indicator in the "normal state" dataset and the mean in the "abnormal state" dataset. The larger this value, the greater the deviation of the indicator when the state changes.

[0063] Within-group variance: refers to the degree of dispersion (usually measured by variance) of a monitoring indicator within its respective datasets under "normal" or "abnormal" conditions. A larger value indicates greater natural fluctuation in the indicator itself.

[0064] Contribution Degree: A defined quantitative output result, calculated using an within-group and between-group distance model, representing the magnitude of an individual monitoring indicator's impact on this emergency. This value can be further normalized to a percentage for easier comparison and ranking.

[0065] Historical Normal State Dataset: This refers to baseline data used for comparison with the current "abnormal state". This dataset is not arbitrary historical data, but rather a collection of traffic flow data that was classified as "normal" during historical periods that had similar conditions to the current time of the anomalous event (such as the same weekday type, the same time of day, and similar weather).

[0066] Comprehensive State Index: This refers to a composite index (such as the congestion index) used to comprehensively assess the current traffic operation status. This index can be a weighted fusion of multiple basic monitoring indicators, or it can be composed of anomaly scores output by a machine learning model, serving as the basis for triggering attribution analysis.

[0067] DBSCAN (Density-Based Spatial Clustering of Applications with Noise) is a density-based spatial clustering algorithm. It can identify clusters of arbitrary shapes and effectively handle noisy data.

[0068] Kalman filtering is an algorithm used to predict and estimate the optimal state of a dynamic system from a series of noisy observations.

[0069] Figure 1 A schematic diagram of the multi-indicator attribution analysis method for traffic anomalies based on intra-group and inter-group distance, according to an embodiment of the present invention, is provided. Figure 1 The traffic anomaly multi-index attribution analysis method based on intra-group and inter-group distance in this embodiment may include the following steps S101 to S103.

[0070] The specific implementation process of steps S101 to S103 is as follows:

[0071] Step S101: Establish cross-frame vehicle trajectories based on point cloud data of the set road segment to obtain continuous motion information of each vehicle, and then calculate multi-dimensional monitoring indicators in real time; the multi-dimensional monitoring indicators include average vehicle speed, traffic flow, time occupancy rate, average headway, and speed difference between lanes.

[0072] In one or more embodiments, a multi-beam lidar (such as 16-line, 32-line, or higher) deployed on highway roadside pillars or gantries is used as the core sensing device to achieve all-weather, high-precision, three-dimensional perception of traffic flow. The specific implementation steps are as follows:

[0073] Step S1011: Acquisition of raw point cloud data;

[0074] The lidar continuously emits a laser beam at a set frequency (e.g., 10Hz) and receives reflected signals from the road surface, vehicles, and other objects, generating a raw point cloud data stream containing three-dimensional coordinates (X, Y, Z), reflection intensity, and timestamps.

[0075] Step S1012: Point cloud preprocessing;

[0076] The original point cloud is filtered and denoised (to remove interference points such as rain, snow, and birds) and the ground is segmented to separate the dynamic target point cloud clusters located above the road surface.

[0077] Step S1013: Vehicle target detection and tracking;

[0078] Clustering algorithms (such as DBSCAN) are used to segment dynamic point cloud clusters and identify individual vehicle targets. Subsequently, Kalman filtering or multi-target tracking algorithms (such as SORT, DeepSORT point cloud-adapted versions) are used to establish cross-frame vehicle trajectories and obtain continuous motion information for each vehicle.

[0079] Step S1014: Calculation of multidimensional monitoring indicators;

[0080] Based on the vehicle trajectory data obtained from tracking, the following core monitoring indicators are calculated in real time within a set road segment analysis unit (e.g., a 500-meter road segment) and time window (e.g., 1 minute):

[0081] average speed The arithmetic mean of the instantaneous speeds of all tracked vehicles within the statistical window;

[0082] ；

[0083] Let be the instantaneous speed of the i-th vehicle; N is the total number of valid vehicles passing through the road segment within this time window;

[0084] Traffic volume: The total number of vehicles passing through the downstream section of a road segment per unit of time.

[0085] Volume = N;

[0086] N represents the number of vehicles that successfully pass through the downstream detection section within a unit of time (e.g., 1 minute).

[0087] Time Occupancy: Calculates the change in the proportion of space occupied by a vehicle within the detection range of the lidar over time, and obtains the time proportion by integration.

[0088] ;

[0089] The time it takes for the i-th vehicle to completely occupy a certain spatial position; T is the total observation time (e.g., 60 seconds).

[0090] Average Time Headway (THW) / Average Distance Gap (ADG): The average time interval or spatial distance between adjacent vehicles in the same lane when they pass a reference section.

[0091] ;

[0092] The moment when the preceding vehicle passes through the cross section; The time when the following vehicle passes through the same cross section; M is the number of valid paired following samples.

[0093] ;

[0094] Let be the spatial distance between the j-th pair of vehicles (in meters).

[0095] Speed ​​difference between lanes: Calculate the average speed of vehicles in each lane, and then find the difference between the maximum and minimum values.

[0096] ;

[0097] Let K be the average speed of all vehicles in the k-th lane; k = 1, 2, ...; K is the total number of lanes (e.g., K = 3 or 4).

[0098] Step S102: Construct and quantify traffic safety indicators based on multi-dimensional monitoring indicators, and calculate comprehensive safety indicators based on the quantified traffic safety indicators and their corresponding weight values; the traffic safety indicators include following risk, lane-changing risk and speed risk.

[0099] By using LiDAR vehicle tracking trajectory data, high-risk driving behaviors are identified and their risk values, i.e., traffic safety indicators, are quantified, including car-following risk, lane-changing risk, and speed risk.

[0100] Specifically, the risk of following another vehicle is characterized by the headway distance (THW) and the time to collision (TTC).

[0101] THW=D / V_ego;

[0102] When V_ego > V_front, TTC = D / (V_ego - V_front) holds true.

[0103] Where D is the distance between the vehicle and the vehicle in front, V_ego is the speed of the vehicle, and V_front is the speed of the vehicle in front.

[0104] Physical meaning and risk interpretation: A low THW indicates insufficient driver reaction time. TTC is a direct measure of the degree of collision approach; the smaller the value, the higher the risk. TTC < 3s is generally considered extremely high risk.

[0105] Specifically, the formula for calculating lane change risk is:

[0106] Lane change risk = α × f(TTC_front) + β × f(TTC_rear);

[0107] When the current vehicle speed is greater than the vehicle in front, TTC_front = d_front / (v_ego - v_front);

[0108] When the speed of the following vehicle is greater than that of this vehicle, TTC_rear = d_rear / (v_rear - v_ego);

[0109] f(TTC) = 1 / TTC;

[0110] Where TTC is the collision time; TTC_front and TTC_rear are the collision times with the vehicles in front and behind in the target lane, respectively; f() is the reciprocal function of the collision time; d_front and d_rear are the distances to the vehicles in front and behind in the target lane, respectively; V_ego is the vehicle speed, and V_front is the speed of the vehicle in front; α and β are the weight values ​​of the corresponding reciprocal functions of the collision time; α + β = 1. The weights of α and β can be adjusted according to the scenario (e.g., on highways, the weight of the vehicle in front is emphasized, and in urban areas, the weight of the vehicle behind is emphasized).

[0111] Specifically, speed risk is assessed using segment-level speed volatility. To represent it, its expression is:

[0112] ;

[0113] ;

[0114] Where M is the total number of valid vehicles passing through the road segment within the time period; This indicates the average acceleration and deceleration intensity of the vehicle; a higher value indicates more unstable driving behavior. Let be the instantaneous speed of the vehicle at time t; Let N be the instantaneous speed of the vehicle at time t-1; N is the number of valid sampling points of the vehicle within window T.

[0115] By aggregating micro-level individual risks within a specific time window (e.g., 1 minute) and spatial range (e.g., a 500-meter road segment), an indicator representing the overall safety level of that road segment is formed.

[0116] Within a time window, count the number of events where TTC < 3s for all vehicles; count the number of events where THW < 1s; count the number of aggressive lane changes exceeding a certain threshold; to avoid bias under different traffic volumes, divide the above counts by the total number of vehicles in that time period to obtain the standardized risk event rate (e.g., the number of high-risk following events per 100 vehicles).

[0117] The comprehensive safety index of the road segment is calculated by weighting using the entropy weight method. The process is as follows:

[0118] Step a: Construct an evaluation matrix; each row represents a time window, and each column represents a traffic safety indicator (following risk, lane-changing risk, and speed risk).

[0119] Step b: Data standardization; eliminating the influence of different dimensions of indicators; calculating information entropy: calculate the information entropy of each indicator. The greater the difference in indicator values ​​(i.e., the greater the fluctuation, the more information it provides), the smaller the entropy; determining weights, the smaller the entropy, the greater the weight of the indicator.

[0120] Step c: Calculate the overall score Safety_Score, which is the overall safety metric;

[0121] Safety_Score= w1*I1 + w2*I2 + w3*I3 + ...;

[0122] Where: I1, I2, I3 are the normalized values ​​of each risk indicator;

[0123] w1, w2, and w3 are the corresponding weights calculated using the entropy weight method.

[0124] Security Status Assessment and Visualization:

[0125] Security levels can be categorized based on long-term historical data, and dynamic thresholds (threshold 1, threshold 2, and threshold 3) can be set to classify security indicators into different levels.

[0126] Safety (can be represented by green): Safety_Score < threshold 1. Traffic flow is smooth and risk events are rare.

[0127] Note (can be indicated in yellow): Threshold 1 ≤ Safety_Score < Threshold 2. Risk events are beginning to occur and require attention.

[0128] Warning (can be indicated in orange): Threshold 2 ≤ Safety_Score < Threshold 3. High-risk events occur frequently, and the probability of accidents increases significantly.

[0129] Danger (can be indicated in red): Safety_Score ≥ 3. The situation is extremely dangerous, with a high risk of traffic accidents.

[0130] Step S103: When abnormal fluctuations in the comprehensive safety index are detected in the current period (e.g., a sudden drop in the comprehensive safety index), the traffic safety index of the current period is compared with that of the previous normal period. By calculating the differences between groups and the degree of dispersion of each traffic safety index within the group, the contribution of each traffic safety index to the change in the comprehensive safety index is quantified, thereby identifying the dominant factors and generating an interpretable attribution report.

[0131] When the overall safety index drops sharply, step S103 above can be executed to quickly locate the specific traffic safety index that caused the sharp drop in the overall safety index (e.g., a surge in the risk of following other vehicles), and analyze the specific reasons, such as whether it is due to large vehicles mixing in, weather conditions, or the traffic accident itself.

[0132] Analysis Dimension: Monthly dimension (M). The same principle applies to the daily dimension (D).

[0133] Data grouping:

[0134] Anomaly group (G_anomaly): Defined as the period in which the indicator fluctuations occur, i.e., all data for this month.

[0135] Normal group (G_normal): defined as all data from the previous month that is most comparable to the current period in the historical period.

[0136] Risk Factor (i): A set of sub-indicators that constitute the comprehensive safety index Safety_Score. For example, when i=1, it is the risk of following other vehicles; when i=2, it is the lane change conflict rate; and when i=3, it is the speed risk.

[0137] The process of quantifying the contribution of each traffic safety indicator to the change in the overall safety index is as follows:

[0138] For each traffic safety indicator, its attribution score is calculated based on the ratio of inter-group variance to intra-group dispersion; the attribution scores of each traffic safety indicator are normalized to obtain its contribution and then ranked accordingly.

[0139] The contribution of each traffic safety indicator to the change in the overall safety index (i.e., the Attribution Score). The formula is:

[0140] ;

[0141] in, , are the mean and variance of the i-th traffic safety indicator in the abnormal group, respectively; , denoted as the mean and variance of the i-th traffic safety indicator in the normal group.

[0142] Numerator (between-group differences): Measures the average deviation of this traffic safety indicator between this month and last month. The larger the value, the more significant the change.

[0143] Denominator (within-group dispersion): Measures the inherent level of fluctuation of the traffic safety indicator within its respective group. The larger the value, the less stable the traffic safety indicator itself is, and the lower the significance of its changes.

[0144] The attribution score of each traffic safety indicator was normalized to obtain its contribution degree. ), and sort them accordingly.

[0145] ;

[0146] Where n is the total number of traffic safety indicators. It directly reflects the contribution ratio of the i-th risk factor to the fluctuation of the overall safety index. The larger the value, the greater the contribution.

[0147] The present invention will be further described below with reference to the embodiments.

[0148] Background: This month's Safety_Score_M1 is 72 (dangerous), while last month's Safety_Score_M0 was 60 (caution), indicating a deterioration of ΔS = +12.

[0149] Table 1. Risk factors and their ranking;

[0150]

[0151] According to Table 1, the main reason for the deterioration of traffic safety this month (ΔS = +12) is the sharp increase in car-following risk, contributing as much as 58.1%. Specifically, the frequency of high-risk car-following events with a TTC < 3 seconds is significantly higher than last month. A secondary reason is the increased speed difference (contributing 25.2%), indicating an increase in sudden acceleration and deceleration in traffic flow. It is recommended to focus on investigating whether there are bottlenecks causing frequent starts and stops in traffic flow this month (such as newly discovered road construction, malfunctioning traffic lights, etc.), and to conduct special rectification of following-to-the-move-in on high-risk road sections.

[0152] The output of this invention is not merely data, but a clear guideline directly pointing to governance actions (e.g., the need to focus on controlling reckless driving), forming a complete closed loop of "monitoring-diagnosis-governance". This invention organically combines micro-level perception, macro-level indicator construction, and advanced attribution models to form a complete, closed-loop, and intelligent traffic safety diagnosis and assessment system, demonstrating significant technological integration and innovation.

[0153] like Figure 2 As shown, the traffic anomaly multi-index attribution analysis system based on intra-group and inter-group distance provided in this embodiment of the invention can be implemented in software. The traffic anomaly multi-index attribution analysis system based on intra-group and inter-group distance includes the following software modules: multi-dimensional monitoring index calculation module 201, comprehensive safety index calculation module 202, and index attribution analysis module 203.

[0154] The following is an introduction to the functions of each software module in the multi-indicator attribution analysis system for traffic anomalies based on intra-group and inter-group distance:

[0155] The multi-dimensional monitoring index calculation module 201 is used to establish cross-frame vehicle trajectories based on point cloud data of a set road segment, obtain continuous motion information of each vehicle, and then calculate multi-dimensional monitoring indicators in real time; the multi-dimensional monitoring indicators include average vehicle speed, traffic flow, time occupancy rate, average headway, and speed difference between lanes.

[0156] The comprehensive safety index calculation module 202 is used to construct and quantify traffic safety indicators based on multi-dimensional monitoring indicators, and calculate comprehensive safety indicators based on the quantified traffic safety indicators and their corresponding weight values; the traffic safety indicators include following risk, lane changing risk and speed risk;

[0157] The indicator attribution analysis module 203 is used to compare the traffic safety indicators of the current period with those of the previous normal period when abnormal fluctuations in the comprehensive safety indicators are detected in the current period. By calculating the differences between groups and the degree of dispersion of each traffic safety indicator within the group, the contribution of each traffic safety indicator to the change of the comprehensive safety indicators is quantified, thereby identifying the dominant factors and generating an interpretable attribution report.

[0158] It should be noted that each module in the traffic anomaly multi-index attribution analysis system based on intra-group and inter-group distance in this embodiment of the invention corresponds one-to-one with each step in the traffic anomaly multi-index attribution analysis method based on intra-group and inter-group distance in the above embodiment, and their specific implementation processes are the same, so they will not be repeated here.

[0159] The structure of the electronic device according to an embodiment of the present invention will be described in detail below. Figure 3 This is a schematic diagram of the composition structure of an electronic device provided in an embodiment of the present invention. It can be understood that... Figure 3 The diagram shows only an exemplary structure of the electronic device, not the entire structure. Some or all of the structures shown may be implemented as needed.

[0160] The electronic device provided in this embodiment of the invention includes: at least one processor 301, a memory 302, a user interface 303, and at least one network interface 304. The various components in the traffic anomaly multi-index attribution analysis system based on intra-group and inter-group distance are coupled together via a bus system 305. It can be understood that the bus system 305 is used to realize the connection and communication between these components. In addition to a data bus, the bus system 305 also includes a power bus, a control bus, and a status signal bus. However, for clarity, in... Figure 3 The general designated all buses as Bus System 305.

[0161] The user interface 303 may include a monitor, keyboard, mouse, trackball, click wheel, buttons, touchpad, or touch screen.

[0162] It is understood that memory 302 can be volatile memory or non-volatile memory, or both. In this embodiment of the invention, memory 302 is capable of storing data to support the operation of the terminal. Examples of this data include any computer programs used to operate on the terminal, such as operating systems and applications. The operating system includes various system programs, such as framework layers, core library layers, driver layers, etc., used to implement various basic services and handle hardware-based tasks. Applications can include various applications.

[0163] In some embodiments, the traffic anomaly multi-index attribution analysis system based on intra-group and inter-group distance provided in this invention can be implemented using a combination of hardware and software. As an example, the system can be a processor in the form of a hardware decoding processor, programmed to execute the traffic anomaly multi-index attribution analysis method based on intra-group and inter-group distance provided in this invention. For example, the hardware decoding processor can employ one or more application-specific integrated circuits (ASICs), DSPs, programmable logic devices (PLDs), complex programmable logic devices (CPLDs), field-programmable gate arrays (FPGAs), or other electronic components.

[0164] As an example, processor 301 can be an integrated circuit chip with signal processing capabilities, such as a general-purpose processor, a digital signal processor (DSP), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc., wherein the general-purpose processor can be a microprocessor or any conventional processor, etc.

[0165] As an example of the hardware implementation of the traffic anomaly multi-index attribution analysis system based on intra-group and inter-group distance provided in this embodiment of the invention, the device provided in this embodiment of the invention can be directly executed by a processor 301 in the form of a hardware decoding processor. For example, it can be executed by one or more application-specific integrated circuits (ASICs), DSPs, programmable logic devices (PLDs), complex programmable logic devices (CPLDs), field-programmable gate arrays (FPGAs), or other electronic components to implement the traffic anomaly multi-index attribution analysis method based on intra-group and inter-group distance provided in this embodiment of the invention.

[0166] The memory 302 in this embodiment of the invention is used to store various types of data to support the operation of a multi-indicator attribution analysis system for traffic anomalies based on intra-group and inter-group distances, or to store data for execution. Figure 1The program code for the method shown. Examples of this data include: any executable instructions for operating on a traffic anomaly multi-index attribution analysis system based on intra-group and inter-group distance, such as executable instructions, and programs implementing the traffic anomaly multi-index attribution analysis method based on intra-group and inter-group distance of this invention can be included in the executable instructions.

[0167] Specifically, according to embodiments of this application, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of this application include a computer program product comprising a computer program carried on a computer-readable medium, the computer program including functions for executing... Figure 1 The program code for the method shown. In such an embodiment, the computer program can be downloaded and installed from a network via a communication component, and / or installed from a removable medium. When the computer program is executed by the central processing unit, it performs the various functions defined in the apparatus of this application.

[0168] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, as well as combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0169] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A multi-indicator attribution analysis method for traffic anomalies based on within-group and between-group distances, characterized in that, include: Based on the point cloud data of the set road section, vehicle trajectories across frames are established to obtain continuous motion information of each vehicle, and then multi-dimensional monitoring indicators are calculated in real time. The multidimensional monitoring indicators include average vehicle speed, traffic flow, time occupancy, average headway, and speed difference between lanes. Based on multidimensional monitoring indicators, traffic safety indicators are constructed and quantified. Based on the quantified traffic safety indicators and their corresponding weight values, a comprehensive safety index is calculated. The traffic safety indicators include following risk, lane-changing risk, and speed risk. When abnormal fluctuations in the overall safety index are detected in the current period, the traffic safety index of the current period is compared with that of the previous normal period. By calculating the differences between groups and the degree of dispersion of each traffic safety index within groups, the contribution of each traffic safety index to the change of the overall safety index is quantified, thereby identifying the dominant factors and generating an interpretable attribution report. The process of quantifying the contribution of each traffic safety indicator to the change in the overall safety indicator is as follows: For each traffic safety indicator, its attribution score is calculated based on the ratio of inter-group variance to intra-group dispersion. The attribution scores of each traffic safety indicator are normalized to obtain their contribution and then ranked accordingly. Contribution of each traffic safety indicator to the change in the overall safety index The formula is: ； in, , are the mean and variance of the i-th traffic safety indicator in the abnormal group, respectively; , denoted as the mean and variance of the i-th traffic safety indicator in the normal group.

2. The multi-indicator attribution analysis method for traffic anomalies based on intra-group and inter-group distance as described in claim 1, characterized in that, By calculating the information entropy of each traffic safety indicator, the corresponding weight of each traffic safety indicator is determined.

3. The multi-indicator attribution analysis method for traffic anomalies based on intra-group and inter-group distance as described in claim 1, characterized in that, Following-the-vehicle risk is characterized by headway distance (THW) and time to collision (TTC). THW=D / V_ego; When V_ego > V_front, TTC = D / (V_ego - V_front) holds true. Where D is the distance between the vehicle and the vehicle in front, V_ego is the speed of the vehicle, and V_front is the speed of the vehicle in front.

4. The multi-indicator attribution analysis method for traffic anomalies based on intra-group and inter-group distance as described in claim 1, characterized in that, The formula for calculating lane change risk is: Lane change risk = α × f(TTC_front) + β × f(TTC_rear); When the current vehicle speed is greater than the vehicle in front, TTC_front = d_front / (v_ego - v_front); When the speed of the following vehicle is greater than that of this vehicle, TTC_rear = d_rear / (v_rear - v_ego); f(TTC) = 1 / TTC; Where TTC is the collision time; TTC_front and TTC_rear are the collision times with the vehicles in front and behind in the target lane, respectively; f() is the reciprocal function of the collision time; d_front and d_rear are the distances to the vehicles in front and behind in the target lane, respectively; V_ego is the vehicle speed, and V_front is the speed of the vehicle in front; α and β are the weight values ​​of the corresponding reciprocal functions of the collision time; α+β =1.

5. The multi-indicator attribution analysis method for traffic anomalies based on intra-group and inter-group distance as described in claim 1, characterized in that, Speed ​​risk is assessed using segment-level speed volatility. To represent it, its expression is: ； ； Where M is the total number of valid vehicles passing through the road segment within the time period; This indicates the average acceleration and deceleration intensity of the vehicle; a higher value indicates more unstable driving behavior. Let be the instantaneous speed of the vehicle at time t; Let N be the instantaneous speed of the vehicle at time t-1; N is the number of valid sampling points of the vehicle within window T.

6. A multi-indicator attribution analysis system for traffic anomalies based on intra-group and inter-group distance, characterized in that, The traffic anomaly multi-indicator attribution analysis method based on intra-group and inter-group distance as described in any one of claims 1-5 includes: The multi-dimensional monitoring index calculation module is used to establish cross-frame vehicle trajectories based on point cloud data of a set road segment, obtain continuous motion information of each vehicle, and then calculate multi-dimensional monitoring indicators in real time; the multi-dimensional monitoring indicators include average vehicle speed, traffic flow, time occupancy rate, average headway, and speed difference between lanes. The comprehensive safety index calculation module is used to construct and quantify traffic safety indicators based on multi-dimensional monitoring indicators, and to calculate the comprehensive safety index based on the quantified traffic safety indicators and their corresponding weight values; the traffic safety indicators include following risk, lane-changing risk and speed risk; The indicator attribution analysis module is used to compare the traffic safety indicators of the current period with those of the previous normal period when abnormal fluctuations in the comprehensive safety indicators are detected in the current period. By calculating the differences between groups and the degree of dispersion of each traffic safety indicator within the group, the module quantifies the contribution of each traffic safety indicator to the change in the comprehensive safety indicators, thereby identifying the dominant factors and generating an interpretable attribution report.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the steps in the multi-indicator attribution analysis method for traffic anomalies based on intra-group and inter-group distance as described in any one of claims 1-5.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the steps in the multi-indicator attribution analysis method for traffic anomalies based on intra-group and inter-group distance as described in any one of claims 1-5.

Images