Method and device for analyzing dynamic behaviors of vehicles on road on highway and medium

Abstract

The invention discloses an expressway in-transit vehicle dynamic behavior analysis method and device and a medium, and belongs to the field of electric digital data processing. The method comprises the following steps: preprocessing multivariate data collected in real time, and fusing the preprocessed multivariate data to form original path information of a vehicle; on the basis of a road network topological structure and a shortest path algorithm, original path information is fitted, and a continuous and complete vehicle driving track is generated; based on the vehicle driving track, comparing the multivariate data in real time, determining whether the vehicle has an abnormal behavior, if yes, continuing to judge an abnormal type corresponding to the vehicle, and generating early warning information corresponding to the abnormal type; and according to the multivariate data of the to-be-predicted road section in the expressway, predicting the traffic flow of the to-be-predicted road section in real time, and displaying the traffic flow and early warning information of the to-be-predicted road section through a user interaction interface, thereby realizing dynamic behavior analysis of the in-transit vehicles on the expressway.

Description

Technical Field

[0001] The present application relates to the technical field of electronic digital data processing, and in particular to a method, device and medium for analyzing the dynamic behavior of vehicles on a highway. Background Art

[0002] At present, with the completion of the national highway network system, the intelligent transportation system faces the dual challenges of massive data processing and precise management. At present, highway toll collection and vehicle behavior monitoring mainly rely on ETC gantry transaction data and license plate recognition data, but due to the high complexity of the road network, equipment recognition errors and data island problems, the integrity and accuracy of vehicle driving path restoration are difficult to guarantee. For example, when a vehicle passes through the gantry, transaction data may be lost due to signal interference, or the license plate recognition device may misjudge the license plate information, resulting in breakpoints or deviations in path fitting.

[0003] In the existing technology, path restoration is mostly based on a single data source, such as ETC transaction flow, and lacks multi-source data fusion and dynamic correction mechanisms, making it difficult to deal with scenarios where gantry data is missing or erroneous. In addition, the real-time detection capability of abnormal behaviors is insufficient, such as blocking ETC equipment, switching passes, etc., and the reliance on manual rules for auditing afterwards is inefficient, which can easily cause toll loss and cannot dynamically combine multi-source data to improve judgment accuracy. In addition, the existing system lacks real-time tracking of vehicles on the road, and traffic flow predictions are mostly based on historical statistics. It is impossible to dynamically adjust the prediction model in combination with real-time path data, resulting in inefficient resource allocation on congested sections. Summary of the invention

[0004] The embodiments of the present application provide a method, device and medium for analyzing the dynamic behavior of vehicles on a highway to solve the above-mentioned technical problems.

[0005] On the one hand, an embodiment of the present application provides a method for analyzing dynamic behavior of vehicles on a highway, including: Preprocess the multivariate data collected in real time and fuse the preprocessed multivariate data to form the original path information of the vehicle; Based on the road network topology and the shortest path algorithm, the original path information is fitted to generate a continuous and complete vehicle driving trajectory; Based on the vehicle's driving trajectory, the multivariate data is compared in real time to determine whether the vehicle has abnormal behavior. If so, the abnormal type corresponding to the vehicle is further determined, and warning information corresponding to the abnormal type is generated; Based on the multivariate data of the section to be predicted in the highway, the traffic flow of the section to be predicted is predicted in real time, and the traffic flow and warning information of the section to be predicted are displayed through the user interaction interface to realize the dynamic behavior analysis of vehicles on the highway.

[0006] In one implementation of the present application, the multivariate data collected in real time is preprocessed, and the preprocessed multivariate data is fused to form the original path information of the vehicle, specifically including: The sensors installed on the highway collect multivariate data of vehicles on each road section in real time; wherein the multivariate data includes ETC gantry transaction data, ETC gantry plate recognition data, toll station entrance and exit transaction data and overweight control data; Cleaning and fusing the collected multivariate data; wherein the cleaned data includes ETC gantry reverse mark data, outliers and abnormal spatiotemporal data; According to the license plate number, vehicle unique identification and travel time, the cleaned and fused multivariate data are associated, and the original path information of the vehicle sorted by time is generated.

[0007] In one implementation of the present application, the collected multivariate data is cleaned and integrated, specifically including: According to the special situation identifier in the ETC gantry transaction data, determine the ETC gantry reverse mark data in the multivariate data, and filter out the ETC gantry reverse mark data; Identify abnormal transaction time or abnormal license plate recognition data in the multivariate data after filtering out the ETC gantry reverse mark data, and filter the abnormal transaction time or abnormal license plate recognition data; wherein the abnormal transaction time is used to indicate that the transaction time is ahead of or greater than a configurable time threshold, and the abnormal license plate recognition data is used to indicate that the recognition license plate credibility of the ETC gantry license plate recognition data is zero, the recognition license plate is empty, or the recognition license plate is a default license plate; According to the ETC gantry license plate recognition data, the distance between adjacent gantries and the time difference of vehicles at adjacent gantries are calculated, and based on the distance between adjacent gantries and the time difference, the abnormal time and space license plate recognition data that exceeds the preset speed threshold is determined to eliminate the abnormal time and space license plate recognition data.

[0008] In one implementation of the present application, the original path information is fitted based on the road network topology and the shortest path algorithm to generate a continuous and complete vehicle driving trajectory, specifically including: The ETC gantries, toll booths and billing units in the highway are abstracted as nodes in a directed graph, and directed edges are formed between two ordered nodes to construct the highway network topology. According to the upstream and downstream relationship of the ETC gantry on the expressway where the vehicle passes, determine whether the ETC gantry is an invalid node; wherein the invalid node includes a node with sudden change in direction, a speed exceeding the limit or an isolated node; According to the shortest path algorithm and the adjacent relationship between nodes, the missing gantry information of the path between the two ETC gantries is filled in to achieve the fitting of the original path information and generate a continuous and complete vehicle driving trajectory.

[0009] In one implementation of the present application, based on the vehicle's driving trajectory, the multivariate data is compared in real time to determine whether the vehicle has abnormal behavior. If so, the abnormal type corresponding to the vehicle is further determined, specifically including: According to the vehicle driving trajectory, the ETC gantry transaction data and the ETC gantry plate recognition data in the multivariate data are compared to obtain a comparison result; According to the comparison result, it is determined that the vehicle has abnormal behavior, and according to the ETC gantry transaction data and the ETC gantry license plate recognition data, it is determined that the abnormal type corresponding to the abnormal behavior of the vehicle is shielding of the passage medium, entry without exit, truck dropping off and replacing trailer, different passage media for entry and exit, or U / J type vehicle behavior.

[0010] In one implementation of the present application, based on the ETC gantry transaction data and the ETC gantry plate recognition data, it is determined that the abnormal type corresponding to the abnormal behavior of the vehicle is shielding the passage medium, entering but not exiting, truck dropping and replacing the trailer, different passage media for the entrance and exit, or U / J type vehicle behavior, specifically including: If the comparison result is that multiple consecutive ETC gantries have gantry brand identification flow but no gantry transaction flow, it is determined that the abnormal type corresponding to the abnormal behavior is shielding of the passage medium; If the comparison result is that the vehicle has an entry pass record but no exit payment flow, then the abnormal type corresponding to the abnormal behavior is determined to be entry without exit; wherein the exit payment flow includes the toll station exit payment flow, the toll station exit supplementary payment flow and the toll station exit outflow flow; If the comparison result is that the difference between the number of axles and the weight at the toll station exit and the toll station entrance in the same trip is greater than the deviation threshold, then it is determined that the abnormal type corresponding to the abnormal behavior is a truck drop-and-hook replacement; If the comparison result is that the type of the medium passing through the inlet water flow is different from the type of the medium passing through the outlet water flow, then it is determined that the abnormal type corresponding to the abnormal behavior is that the medium passing through the inlet and outlet is different; If the comparison result is that the vehicle has detoured and triggered the minimum fee threshold at the exit, it is determined that the abnormal type corresponding to the abnormal behavior is U / J type vehicle behavior.

[0011] In one implementation of the present application, based on the multi-dimensional data of the road section to be predicted in the highway, the traffic flow of the road section to be predicted is predicted in real time, specifically including: Build a vehicle pool for ETC gantries and toll stations, encode the real-time ETC gantry transaction data and toll station inflow and outflow data by node, store them in the vehicle pool, and record the vehicle location and travel time; After the vehicle passes through the ETC gantry or toll station, the ETC gantry transaction data and the inflow and outflow of the toll station are removed from the upstream node vehicle pool and added to the current node vehicle pool; wherein the upstream node of the ETC gantry includes the upstream toll station and the upstream gantry, and the upstream node of the toll station includes the upstream upstream gantry and the downstream upstream gantry; The gantry traffic flow is calculated based on the traffic flow of the upstream gantry, the traffic flow at the toll station entrance, the traffic flow at the toll station exit and the current gantry traffic flow, and the toll station traffic flow is calculated based on the upstream gantry traffic flow, the upstream downstream gantry traffic flow, the downstream upstream gantry traffic flow and the downstream downstream gantry traffic flow.

[0012] In one implementation of the present application, after displaying the traffic flow and warning information of the road section to be predicted through the user interaction interface, the method further includes: Highlight vehicles with abnormal behavior and display the corresponding abnormal type, location information and handling suggestions on the user interaction interface; For vehicles with disputed highway tolls, the billing path and the actual fitting path are synchronized, and the difference area between the billing path and the actual fitting path is displayed; According to the time period and road section range, the predicted road section traffic flow and abnormal behavior statistics are displayed, and correction instructions for the fitted path are received to optimize the algorithm parameters of the shortest path algorithm.

[0013] On the other hand, the embodiment of the present application further provides a device for analyzing dynamic behavior of vehicles on a highway, the device comprising: at least one processor; and, a memory communicatively coupled to the at least one processor; The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the above-mentioned method for analyzing the dynamic behavior of vehicles on a highway.

[0014] On the other hand, an embodiment of the present application further provides a non-volatile computer storage medium storing computer executable instructions, which, when executed, implements a method for analyzing dynamic behavior of vehicles on a highway as described above.

[0015] The present application provides a method, device and medium for analyzing dynamic behavior of vehicles on a highway, which at least have the following beneficial effects: By collecting multivariate data in real time and preprocessing it, noise and redundant information can be effectively removed, data accuracy and availability can be improved, and information from different sources can be integrated to form more comprehensive and accurate original vehicle path information; fitting the original path information based on the road network topology and the shortest path algorithm can generate continuous and complete vehicle driving trajectories, which helps to track the vehicle driving process more accurately and provide a reliable basis for abnormal behavior detection and traffic flow prediction; by comparing multivariate data in real time, abnormal behaviors of vehicles can be discovered in time and the type of abnormality can be quickly determined; generating corresponding warning information can help management departments take timely measures to prevent potential safety hazards and traffic violations and improve road safety; predicting traffic flow in real time based on multivariate data of the road section to be predicted can provide timely traffic status information for traffic management departments, help optimize traffic flow management, reduce congestion and delays, and improve road traffic efficiency; displaying the traffic flow and warning information of the road section to be predicted through the user interaction interface makes the information more intuitive and easy to understand, which not only improves the user experience, but also provides decision-making support for management departments and helps to formulate more scientific and reasonable traffic management strategies. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings: Figure 1 A schematic diagram of a flow chart of a method for analyzing dynamic behavior of vehicles on a highway provided in an embodiment of the present application; Figure 2 A flowchart of another method for analyzing dynamic behavior of vehicles on a highway provided in an embodiment of the present application; Figure 3 A schematic diagram of the internal structure of a device for analyzing the dynamic behavior of vehicles on a highway provided in an embodiment of the present application. DETAILED DESCRIPTION

[0017] In order to make the purpose, technical solution and advantages of the present application clearer, the technical solution of the present application will be clearly and completely described below in combination with the specific embodiments of the present application and the corresponding drawings. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present application.

[0018] The technical solutions provided by various embodiments of the present application are described in detail below in conjunction with the accompanying drawings.

[0019] Figure 1A flowchart of a method for analyzing dynamic behavior of vehicles on a highway provided in an embodiment of the present application.

[0020] The analysis method involved in the embodiments of the present application can be implemented by a terminal device or a server, and the present application does not impose any special restrictions on this. For the convenience of understanding and description, the following embodiments are described in detail by taking a server as an example.

[0021] It should be noted that the server may be a single device or a system consisting of multiple devices, that is, a distributed server, and this application does not make any specific limitation on this.

[0022] like Figure 1 As shown, a method for analyzing dynamic behavior of vehicles on a highway provided by an embodiment of the present application includes: 101. Preprocess the multivariate data collected in real time, and fuse the preprocessed multivariate data to form the original path information of the vehicle.

[0023] Collect the traffic flow data of each vehicle passing on the highway in real time, including ETC gantry transaction data, license plate recognition data, entrance flow data, exit flow data, and overload control data. Pre-process the traffic flow data to remove invalid interference data; associate and fuse the vehicle traffic flow by time according to the license plate and vehicle traffic identification to form the original path information sorted by time.

[0024] Specifically, in one embodiment of the present application, the multivariate data collected in real time is preprocessed, and the preprocessed multivariate data is fused to form the original path information of the vehicle, which specifically includes: Sensors installed on the highway collect multi-dimensional data of vehicles on each road section in real time; the multi-dimensional data includes ETC gantry transaction data, ETC gantry plate recognition data, toll station entrance and exit transaction data, and overweight control data; Clean and fuse the collected multivariate data; the cleaned data includes ETC gantry reverse mark data, outliers, and abnormal spatiotemporal data; According to the license plate number, vehicle unique identification and travel time, the cleaned and fused multivariate data are associated, and the original path information of the vehicle sorted by time is generated.

[0025] In one embodiment, ETC gantry transaction data, license plate recognition data, toll station entry and exit transaction data, and overweight control data are collected in real time. The collected data is further processed to form the original path information of vehicle passage through data cleaning and multivariate data fusion.

[0026] Data cleaning includes compliance verification of the collected data, such as ETC gantry ETC gantry reverse label data, outlier filtering, time and space verification, etc. to eliminate abnormal data and ensure data accuracy.

[0027] In one embodiment of the present application, cleaning and fusing the collected multi-dimensional data specifically includes: According to the special situation identification in the ETC gantry transaction data, the ETC gantry reverse label data in the multivariate data is determined, and the ETC gantry reverse label data is filtered out; Identify abnormal transaction time or abnormal license plate recognition data in the multivariate data after filtering out the ETC gantry reverse mark data, and filter the abnormal transaction time or abnormal license plate recognition data; wherein, the abnormal transaction time is used to indicate that the transaction time is ahead of or greater than a configurable time threshold, and the abnormal license plate recognition data is used to indicate that the recognition license plate credibility of the ETC gantry license plate recognition data is zero, the recognition license plate is empty, or the recognition license plate is a default license plate; Based on the ETC gantry license plate recognition data, the distance between adjacent gantries and the time difference of vehicles at adjacent gantries are calculated, and based on the distance and time difference between adjacent gantries, the abnormal time and space license plate recognition data that exceeds the preset speed threshold is determined to eliminate the abnormal time and space license plate recognition data.

[0028] In one embodiment, the reverse mark data of the ETC gantry is excluded, and based on the special situation identification judgment of the ETC gantry transaction flow, abnormal data such as special situations 154-reverse interference, 186-reverse interference, etc. are screened out.

[0029] Filter outliers to exclude situations where the transaction time is ahead of or greater than the current time (configurable time threshold), the ETC gantry license plate recognition data has a zero credibility, the recognized license plate is empty, the recognized license plate is the default license plate, etc.

[0030] Perform time-space verification. If the ETC gantry license plate recognition data has errors in identifying license plates, it is necessary to perform time-space verification. Combined with the time and gantry position information, verify the validity of the ETC gantry license plate recognition data within a specific time and space range. The time difference between the capture time of the vehicle's upstream gantry license plate recognition data and the capture time of the current gantry license plate recognition, and the distance between the gantries determine whether the gantry license plate recognition data is reasonable.

[0031] The gantry vehicle information of the vehicle is added to the vehicle pool of the gantry as the basis for downstream nodes to predict the traffic flow. The collected multi-dimensional data are associated and fused through conditions such as license plate number, vehicle unique identification, and travel time to form the original path information sorted by time.

[0032] 102. Based on the road network topology and the shortest path algorithm, the original path information is fitted to generate a continuous and complete vehicle driving trajectory.

[0033] Specifically, in one embodiment of the present application, based on the road network topology and the shortest path algorithm, the original path information is fitted to generate a continuous and complete vehicle driving trajectory, which specifically includes: The ETC gantries, toll booths and billing units in the highway are abstracted as nodes in a directed graph, and directed edges are formed between two ordered nodes to construct the highway network topology. According to the upstream and downstream relationship of the ETC gantry on the expressway where the vehicle passes, determine whether the ETC gantry is an invalid node; invalid nodes include nodes with sudden changes in direction, speed exceeding the limit or isolated nodes; According to the shortest path algorithm and the adjacent relationship between nodes, the missing gantry information of the path between two ETC gantries is filled in to fit the original path information and generate a continuous and complete vehicle driving trajectory.

[0034] In one embodiment, based on basic data such as the connectivity of ETC gantries, toll stations, and billing units within the province, the province's toll infrastructure is abstractly drawn as a directional road network topology map, with ETC gantries, toll stations, and billing units as points. Directed edges are formed between two ordered points to construct vehicle travel paths, providing a basic information base for the path restoration algorithm.

[0035] The cleaned in-transit data, including toll station entrance flow, toll station exit flow, ETC gantry transaction flow, and ETC gantry plate recognition flow, are used. The toll station entrance flow is used as the starting point, and the toll station exit flow is used as the end point. The ETC gantry transaction flow and the gantry plate recognition flow are fused in time order as the passing point, and the accurate passage trajectory of the vehicle from the entrance station to the exit station is obtained by combining the road network topology map and the path algorithm fitting restoration.

[0036] The gantry nodes that the vehicle passes through should be orderly and continuous. During the path fitting process, according to the directionality between nodes, by recording the previous node and transaction time of each gantry node, it is determined whether the gantry node is an invalid node, such as a sudden change in node direction, whether the speed calculated by the distance / time of a node and the previous and next nodes exceeds the reasonable range. If the node position offset is large, it may cause the cross node to be unable to connect correctly, and thus be identified as an invalid node. Through the topological structure of the road network, isolated nodes or unreachable nodes are identified. At the same time, according to the node adjacent relationship and the shortest path planning, the missing gantry node information between the two gantry nodes is completed, and the integrity and continuity of the path are ensured by deleting invalid nodes in the path and filling the missing nodes in the path.

[0037] After path fitting, a complete and continuous vehicle driving path will be obtained, which includes the main information of the original flow of vehicle driving, transaction time, amount information, node information, etc., as well as the gantry information restored by fitting.

[0038] 103. Based on the vehicle's driving trajectory, multivariate data is compared in real time to determine whether the vehicle has abnormal behavior. If so, the abnormal type corresponding to the vehicle is further determined, and warning information corresponding to the abnormal type is generated.

[0039] Specifically, in one embodiment of the present application, based on the vehicle's driving trajectory, multivariate data is compared in real time to determine whether the vehicle has abnormal behavior. If so, the abnormal type corresponding to the vehicle is further determined, specifically including: According to the vehicle driving trajectory, the ETC gantry transaction data and the ETC gantry plate recognition data in the multivariate data are compared to obtain the comparison result; Based on the comparison results, it is determined that the vehicle has abnormal behavior, and based on the ETC gantry transaction data and the ETC gantry license plate recognition data, it is determined that the abnormal type corresponding to the vehicle's abnormal behavior is blocked traffic medium, entry without exit, truck drop-off and replacement of trailers, different traffic media for entry and exit, or U / J type vehicle behavior.

[0040] In one embodiment, a comparison and analysis is performed based on the collected multi-dimensional data while the vehicle is driving, and a real-time determination is made as to whether the vehicle is blocking the traffic medium, has entry but no exit, has a truck dropping or changing trailers, has different traffic media at the entrance and exit, or has U / J-type abnormal behavior, so that abnormal vehicles can be discovered in time and early warnings can be issued and intercepted.

[0041] In one embodiment of the present application, based on the ETC gantry transaction data and the ETC gantry plate recognition data, it is determined that the abnormal type corresponding to the abnormal behavior of the vehicle is shielding the passage medium, entering but not exiting, truck dropping and replacing the trailer, different passage media for the entrance and exit, or U / J type vehicle behavior, specifically including: If the comparison result shows that multiple consecutive ETC gantries have gantry brand identification flow but no gantry transaction flow, then the abnormal type corresponding to the abnormal behavior is determined to be shielded passage medium; If the comparison result shows that the vehicle has an entry pass record but no exit payment flow, then the abnormal type corresponding to the abnormal behavior is determined to be entry without exit; among which, the exit payment flow includes the toll station exit payment flow, the toll station exit supplementary payment flow and the toll station exit outflow flow; If the comparison result shows that the difference between the number of axles and the weight at the toll station exit and the toll station entrance in the same trip is greater than the deviation threshold, the abnormal type corresponding to the abnormal behavior is determined to be a truck dropping and replacing a trailer; If the comparison result shows that the medium of the inlet flow water is different from the medium of the outlet flow water, then the abnormal type corresponding to the abnormal behavior is determined to be the different medium of the inlet and outlet flow water; If the comparison result shows that the vehicle has detoured and triggered the minimum fee threshold at the exit, it is determined that the abnormal type corresponding to the abnormal behavior is U / J type vehicle behavior.

[0042] In one embodiment, after the path fusion service analyzes the passage of a vehicle, if it is shown that the vehicle has multiple consecutive gantries (the number of gantries can be set) with gantry plate identification flow but no gantry transaction flow, it is considered that there is a suspicion of blocking the passage medium.

[0043] Entry without exit means that the vehicle obtains the normal pass record (such as ETC or CPC card) at the entrance of the expressway, but no payment flow is generated at the exit. It may be due to the following vehicle rushing through the post, opening the road crossing privately, abnormal toll collection software, etc., which leads to the failure to pay the toll. After the path fusion service analysis of the vehicle's passage, it is shown that the vehicle has no provincial boundary exit gantry transaction flow, no provincial boundary exit gantry plate recognition flow, no toll station exit normal flow, no toll station exit supplement flow, and no toll station exit outflow flow, which is considered to be suspected of entry without exit.

[0044] When analyzing the data of a tractor, if the vehicle has more exit axles than entry axles in the same trip or there is a big difference between the entry and exit weights, it is considered as a suspected truck dropout or replacement.

[0045] In the case where the vehicle's inlet and outlet flow media are inconsistent during the same trip, there is a suspicion of card swapping, such as CPC card in and OBU out, or CPC card A in and CPC card B out.

[0046] When the billing amount used by a vehicle in the passage medium is greater than the achievable "minimum fare" path threshold, the "minimum fare" billing operating rules are adopted, and the vehicle detoured within the road network for a long time, and the exit triggers the "minimum fare" charge, thereby achieving the purpose of running long and buying short.

[0047] 104. Based on the multivariate data of the section to be predicted on the highway, the traffic flow of the section to be predicted is predicted in real time, and the traffic flow and warning information of the section to be predicted are displayed through the user interaction interface to realize the dynamic behavior analysis of vehicles on the highway.

[0048] The original path information is processed by the path fitting service, the erroneous nodes in the path are deleted, and the missing nodes in the path are filled to obtain a complete and continuous vehicle driving path, which is then displayed in combination with the front-end GIS map to display the vehicle driving trajectory.

[0049] The vehicle information collected from ETC gantry transactions and toll station transactions is added to the vehicle pool of this gantry. The traffic network topology composed of toll stations and gantries, the road between each node and the downstream node is regarded as a vehicle pool, and each vehicle pool contains vehicles on the way. After the vehicle passes through the node, the vehicle is added to the vehicle pool of this gantry as the basis for the downstream node to predict the traffic flow, and the vehicle data in the upstream node vehicle pool is deleted at the same time.

[0050] Specifically, in one embodiment of the present application, based on the multi-dimensional data of the road section to be predicted in the highway, the traffic flow of the road section to be predicted is predicted in real time, specifically including: Build a vehicle pool for ETC gantries and toll stations, encode the real-time ETC gantry transaction data and toll station inflow and outflow data by node, store them in the vehicle pool, and record the vehicle location and travel time; After the vehicle passes through the ETC gantry or toll station, the ETC gantry transaction data and the inflow and outflow of the toll station are removed from the upstream node vehicle pool and added to the current node vehicle pool; wherein, the upstream node of the ETC gantry includes the upstream toll station and the upstream gantry, and the upstream node of the toll station includes the upstream upstream gantry and the downstream upstream gantry; The gantry traffic flow is calculated based on the traffic flow of the upstream gantry, the traffic flow at the toll station entrance, the traffic flow at the toll station exit and the current gantry traffic flow, and the toll station traffic flow is calculated based on the upstream gantry traffic flow, the upstream downstream gantry traffic flow, the downstream upstream gantry traffic flow and the downstream downstream gantry traffic flow.

[0051] In one embodiment, the vehicle traffic data collected can be used to calculate the number of vehicles on the highway and their information, and to make real-time predictions of the traffic volume on the road section, etc., to provide real-time data support for the traffic management department, helping the management department to reasonably dispatch traffic resources, optimize traffic flow, reduce congestion, and improve road traffic efficiency.

[0052] The traffic network topology consists of toll stations and gantries. Toll stations and gantries are equipped with in-transit vehicle pools. During data preprocessing, the vehicle information of toll station flows and ETC gantry transaction flows is stored according to the toll station code and gantry code. A node traffic flow prediction algorithm is formulated to provide toll station and gantry traffic flow estimation services. The predicted traffic flow information can be provided for front-end display and external interfaces to help management departments reasonably dispatch traffic resources and improve road traffic efficiency.

[0053] The road between each node and the downstream node is considered as a vehicle pool, and each vehicle pool contains vehicles on the way. After a vehicle passes through a node, it is added to the vehicle pool of this gantry as the basis for the downstream node to predict the traffic flow, and the vehicle data in the upstream node vehicle pool is deleted at the same time.

[0054] Without considering the hub interchange, there are generally two upstream nodes of the gantry: upstream toll station and upstream gantry. There are also generally two upstream nodes of the toll station: upstream upstream gantry and downstream upstream gantry. From this, the formula for calculating the traffic flow of the gantry or toll station can be obtained: Gantry traffic flow = upstream gantry traffic flow + (toll station entrance - toll station exit (vehicles passing through the upstream gantry)) - current gantry traffic flow.

[0055] Traffic flow at toll station = upstream gantry - upstream gantry + downstream gantry - downstream gantry.

[0056] In one embodiment of the present application, after displaying the traffic flow and warning information of the road section to be predicted through the user interaction interface, it also includes: Highlight vehicles with abnormal behaviors and display the corresponding abnormality type, location information, and handling suggestions on the user interface; For vehicles with disputed highway tolls, the billing path and the actual fitting path are synchronized, and the difference area between the billing path and the actual fitting path is displayed; According to the time period and road section range, the predicted road section traffic flow and abnormal behavior statistics are displayed, and correction instructions for the fitted path are received to optimize the algorithm parameters of the shortest path algorithm.

[0057] In one embodiment, the location information of the ETC gantries and toll stations in the province and the vehicle driving trajectory are intuitively displayed based on the GIS map. First, for in-transit vehicle inquiries, the in-transit vehicle information of the ETC gantries and toll stations is displayed. The in-transit vehicle information can be queried by license plate number, vehicle model, and traffic node.

[0058] For vehicle historical route query, it provides vehicle historical travel records query by license plate, entry time, and exit time, and restores the vehicle travel trajectory based on the historical records, and renders the path on the map to intuitively display the vehicle's billing path and actual driving trajectory. When there is a toll dispute, it can provide the owner with clear driving path evidence to help resolve the toll dispute. Support the development of related businesses such as audit and data analysis.

[0059] For abnormal vehicle behavior classification query, we provide abnormal driving record query by license plate, time, abnormal behavior, and support simultaneous display of billing path and actual driving path on the map. For road section traffic flow prediction, we provide predicted traffic flow query for road sections, toll booths, and gantries to help management departments reasonably dispatch traffic resources in advance.

[0060] This application can perform real-time path fitting processing on-the-go vehicles, detect abnormal behavior of vehicles in a timely manner, monitor and warn vehicles in real time, make real-time predictions on road traffic flow, provide real-time data support for traffic management departments, and present vehicle driving trajectories to toll collectors and users in a friendly, mapped manner, thereby improving user experience and providing strong support for intelligent management of highways.

[0061] Figure 2 A flowchart of another method for analyzing dynamic behavior of vehicles on a highway provided in an embodiment of the present application. Figure 2As shown, it collects multiple data such as entrances, exits, gantry and overload control, receives these flows, and then performs real-time fusion services on traffic flows, estimates real-time vehicle flow, implements vehicle flow estimation services, and then pushes and stores the estimated vehicle flow in the vehicle pool of vehicles on the way. At the same time, it also needs to compare and complete the paths through traffic paths, implement path inspection services, generate unique flows, and store the unique flows in the preset traffic library.

[0062] In addition, the present application also provides a user interaction interface, and provides a unified query interface on the user interaction interface, including a flow query interface, an in-transit vehicle query, and a toll station traffic flow query.

[0063] The above is an embodiment of the method proposed in this application. Based on the same inventive concept, this application embodiment also provides a highway on-the-way vehicle dynamic behavior analysis device, whose structure is as follows: Figure 3 shown.

[0064] Figure 3 The internal structure diagram of a highway vehicle dynamic behavior analysis device provided in the embodiment of the present application. Figure 3 As shown, the device includes: at least one processor; and, a memory communicatively coupled to the at least one processor; The memory stores instructions that can be executed by at least one processor, and the instructions are executed by at least one processor to enable the at least one processor to: Preprocess the multivariate data collected in real time and fuse the preprocessed multivariate data to form the original path information of the vehicle; Based on the road network topology and the shortest path algorithm, the original path information is fitted to generate a continuous and complete vehicle driving trajectory; Based on the vehicle's driving trajectory, multivariate data is compared in real time to determine whether the vehicle has abnormal behavior. If so, the abnormal type corresponding to the vehicle is determined and warning information corresponding to the abnormal type is generated; Based on the multivariate data of the section to be predicted on the highway, the traffic flow of the section to be predicted is predicted in real time, and the traffic flow and warning information of the section to be predicted are displayed through the user interaction interface to realize the dynamic behavior analysis of vehicles on the highway.

[0065] The present application also provides a non-volatile computer storage medium storing computer executable instructions. When the computer executable instructions are executed, they can: Preprocess the multivariate data collected in real time and fuse the preprocessed multivariate data to form the original path information of the vehicle; Based on the road network topology and the shortest path algorithm, the original path information is fitted to generate a continuous and complete vehicle driving trajectory; Based on the vehicle's driving trajectory, multivariate data is compared in real time to determine whether the vehicle has abnormal behavior. If so, the abnormal type corresponding to the vehicle is determined and warning information corresponding to the abnormal type is generated; Based on the multivariate data of the section to be predicted on the highway, the traffic flow of the section to be predicted is predicted in real time, and the traffic flow and warning information of the section to be predicted are displayed through the user interaction interface to realize the dynamic behavior analysis of vehicles on the highway.

[0066] The above is only an embodiment of the present application and is not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the scope of the claims of the present application.

Claims

1. A method for analyzing dynamic behavior of vehicles on a highway, characterized in that: The method comprises: Preprocess the multivariate data collected in real time and fuse the preprocessed multivariate data to form the original path information of the vehicle; Based on the road network topology and the shortest path algorithm, the original path information is fitted to generate a continuous and complete vehicle driving trajectory; Based on the vehicle's driving trajectory, the multivariate data is compared in real time to determine whether the vehicle has abnormal behavior. If so, the abnormal type corresponding to the vehicle is further determined, and warning information corresponding to the abnormal type is generated; Based on the multivariate data of the section to be predicted in the highway, the traffic flow of the section to be predicted is predicted in real time, and the traffic flow and warning information of the section to be predicted are displayed through the user interaction interface to realize the dynamic behavior analysis of vehicles on the highway.

2. The method for analyzing dynamic behavior of vehicles on a highway according to claim 1, characterized in that: Preprocess the multivariate data collected in real time and fuse the preprocessed multivariate data to form the original path information of the vehicle, including: The sensors installed on the highway collect multivariate data of vehicles on each road section in real time; wherein the multivariate data includes ETC gantry transaction data, ETC gantry plate recognition data, toll station entrance and exit transaction data and overweight control data; Cleaning and fusing the collected multivariate data; wherein the cleaned data includes ETC gantry reverse mark data, outliers and abnormal spatiotemporal data; According to the license plate number, vehicle unique identification and travel time, the cleaned and fused multivariate data are associated, and the original path information of the vehicle sorted by time is generated.

3. A method for analyzing dynamic behavior of vehicles on a highway according to claim 2, characterized in that: Cleaning and fusing the collected multivariate data specifically includes: According to the special situation identifier in the ETC gantry transaction data, determine the ETC gantry reverse mark data in the multivariate data, and filter out the ETC gantry reverse mark data; Identify abnormal transaction time or abnormal license plate recognition data in the multivariate data after filtering out the ETC gantry reverse mark data, and filter the abnormal transaction time or abnormal license plate recognition data; wherein the abnormal transaction time is used to indicate that the transaction time is ahead of or greater than a configurable time threshold, and the abnormal license plate recognition data is used to indicate that the recognition license plate credibility of the ETC gantry license plate recognition data is zero, the recognition license plate is empty, or the recognition license plate is a default license plate; According to the ETC gantry license plate recognition data, the distance between adjacent gantries and the time difference of vehicles at adjacent gantries are calculated, and based on the distance between adjacent gantries and the time difference, the abnormal time and space license plate recognition data that exceeds the preset speed threshold is determined to eliminate the abnormal time and space license plate recognition data.

4. The method for analyzing dynamic behavior of vehicles on a highway according to claim 1, characterized in that: Based on the road network topology and the shortest path algorithm, the original path information is fitted to generate a continuous and complete vehicle driving trajectory, specifically including: The ETC gantries, toll booths and billing units in the highway are abstracted as nodes in a directed graph, and directed edges are formed between two ordered nodes to construct the highway network topology. According to the upstream and downstream relationship of the ETC gantry on the expressway where the vehicle passes, determine whether the ETC gantry is an invalid node; wherein the invalid node includes a node with sudden change in direction, a speed exceeding the limit or an isolated node; According to the shortest path algorithm and the adjacent relationship between nodes, the missing gantry information of the path between the two ETC gantries is filled in to achieve the fitting of the original path information and generate a continuous and complete vehicle driving trajectory.

5. The method for analyzing dynamic behavior of vehicles on a highway according to claim 1, characterized in that: Based on the vehicle's driving trajectory, the multivariate data is compared in real time to determine whether the vehicle has abnormal behavior. If so, the abnormal type corresponding to the vehicle is further determined, specifically including: According to the vehicle driving trajectory, the ETC gantry transaction data and the ETC gantry plate recognition data in the multivariate data are compared to obtain a comparison result; According to the comparison result, it is determined that the vehicle has abnormal behavior, and according to the ETC gantry transaction data and the ETC gantry license plate recognition data, it is determined that the abnormal type corresponding to the abnormal behavior of the vehicle is shielding of the passage medium, entry without exit, truck dropping off and replacing trailer, different passage media for entry and exit, or U / J type vehicle behavior.

6. A method for analyzing dynamic behavior of vehicles on a highway according to claim 5, characterized in that: According to the ETC gantry transaction data and the ETC gantry plate recognition data, the abnormal type corresponding to the abnormal behavior of the vehicle is determined to be shielding of the passage medium, entry without exit, truck drop-off and replacement, different entry and exit passage media or U / J type vehicle behavior, specifically including: If the comparison result is that multiple consecutive ETC gantries have gantry brand identification flow but no gantry transaction flow, it is determined that the abnormal type corresponding to the abnormal behavior is shielding of the passage medium; If the comparison result is that the vehicle has an entry pass record but no exit payment flow, then the abnormal type corresponding to the abnormal behavior is determined to be entry without exit; wherein the exit payment flow includes the toll station exit payment flow, the toll station exit supplementary payment flow and the toll station exit outflow flow; If the comparison result is that the difference between the number of axles and the weight at the toll station exit and the toll station entrance in the same trip is greater than the deviation threshold, then the abnormal type corresponding to the abnormal behavior is determined to be a truck dropping or replacing a trailer; If the comparison result is that the type of the medium passing through the inlet water flow is different from the type of the medium passing through the outlet water flow, then it is determined that the abnormal type corresponding to the abnormal behavior is that the medium passing through the inlet and outlet is different; If the comparison result is that the vehicle has detoured and triggered the minimum fee threshold at the exit, it is determined that the abnormal type corresponding to the abnormal behavior is U / J type vehicle behavior.

7. The method for analyzing dynamic behavior of vehicles on a highway according to claim 1, characterized in that: According to the multivariate data of the road section to be predicted in the highway, the traffic flow of the road section to be predicted is predicted in real time, specifically including: Build a vehicle pool for ETC gantries and toll stations, encode the real-time ETC gantry transaction data and toll station inflow and outflow data by node, store them in the vehicle pool, and record the vehicle location and travel time; After the vehicle passes through the ETC gantry or toll station, the ETC gantry transaction data and the inflow and outflow of the toll station are removed from the upstream node vehicle pool and added to the current node vehicle pool; wherein the upstream node of the ETC gantry includes the upstream toll station and the upstream gantry, and the upstream node of the toll station includes the upstream upstream gantry and the downstream upstream gantry; The gantry traffic flow is calculated based on the traffic flow of the upstream gantry, the traffic flow at the toll station entrance, the traffic flow at the toll station exit and the current gantry traffic flow, and the toll station traffic flow is calculated based on the upstream gantry traffic flow, the upstream downstream gantry traffic flow, the downstream upstream gantry traffic flow and the downstream downstream gantry traffic flow.

8. The method for analyzing dynamic behavior of vehicles on a highway according to claim 1, characterized in that: After displaying the traffic flow and warning information of the road section to be predicted through the user interaction interface, the method further includes: Highlight vehicles with abnormal behavior and display the corresponding abnormal type, location information and handling suggestions on the user interaction interface; For vehicles with disputed highway tolls, the billing path and the actual fitting path are synchronized, and the difference area between the billing path and the actual fitting path is displayed; According to the time period and road section range, the predicted road section traffic flow and abnormal behavior statistics are displayed, and correction instructions for the fitted path are received to optimize the algorithm parameters of the shortest path algorithm.

9. A device for analyzing dynamic behavior of vehicles on a highway, characterized in that: The device comprises: at least one processor; and, a memory communicatively coupled to the at least one processor; Wherein, the memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute a method for analyzing dynamic behavior of vehicles on a highway as described in any one of claims 1-8.

10. A non-volatile computer storage medium storing computer executable instructions, characterized in that: When the computer executable instructions are executed, a method for analyzing dynamic behavior of vehicles on a highway as described in any one of claims 1 to 8 is implemented.

Images