Road traffic analysis method and system based on edge computing

By deploying edge computing devices in the traffic monitoring network to process traffic flow data and performing cloud storage configuration in the traffic data management center, the problems of data verification and storage in the existing technology are solved, and the real-time and accuracy of traffic flow analysis are improved. , realizes effective management and long-term preservation of traffic data.

CN118840872BActive Publication Date: 2024-11-26SICHUAN HUATI LIGHTING TECH
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411321517.0
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-09-23
Publication Date
2024-11-26
Estimated Expiration
2044-09-23

AI Technical Summary

Technical Problem

The existing traffic flow analysis methods based on edge computing lack effective data verification mechanisms, which may lead to invalid or incorrect data being misused, affecting the accuracy of analysis results, and lack of effective storage and management solutions, making it difficult to support long-term data. Analysis and mining work.

Method used

By obtaining real-time road traffic data in the traffic monitoring network, deploying edge computing devices at each traffic observation point for data processing, generating real-time traffic status update descriptions, and cloud storage configuration in the traffic data management center to ensure the effectiveness of data and Long-term preservation. Specific methods include using Gaussian filtering algorithm to remove noise, capturing features in convolutional neural networks, calculating weights of attention mechanisms, semantic coding and full connection mapping to generate traffic congestion features, and performing data verification and storage.

Benefits of technology

Significantly improve the real-time and accuracy of traffic status monitoring, reduce data transmission delay, improve system response speed, ensure data synchronization and integrity, provide a reliable data foundation for traffic management, and achieve effective Long-term preservation and management of edge computing records.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118840872B_ABST
    Figure CN118840872B_ABST
Patent Text Reader

Abstract

The present application provides a road traffic analysis method and system based on edge computing, which obtains road traffic data from multiple traffic observation points in a traffic monitoring network in real time, collects them uniformly when a specific traffic flow update activity is triggered, and uses edge computing to quickly process them near the data source to generate a description reflecting real-time traffic status updates. The key to this method is the instant verification of edge computing results, ensuring that the data of all observation points are valid before storing their edge computing records securely in the cloud of the traffic data management center, which not only improves the real-time and accuracy of traffic flow analysis, but also realizes efficient management and long-term use of data through cloud storage.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the field of computer technology, and more specifically, to a road traffic analysis method and system based on edge computing. Background Art

[0002] With the acceleration of urbanization, urban traffic congestion is becoming increasingly serious. Real-time monitoring and analysis of traffic flow has become an important means to solve traffic problems and optimize traffic management. Traditional traffic flow analysis methods usually rely on a centralized data processing model, that is, data distributed at various traffic monitoring points are first collected to a central server, and then processed and analyzed uniformly by the server. However, this model faces problems such as data transmission delays, high bandwidth usage, and heavy server load when processing massive traffic data, resulting in insufficient real-time performance and difficulty in meeting the needs of modern urban traffic management.

[0003] In order to overcome the limitations of traditional methods, edge computing technology has gradually been introduced into the field of traffic flow analysis. Edge computing significantly reduces the amount of data transmission, reduces network latency, and improves the real-time and efficiency of data processing by processing near the data source. In the traffic monitoring network, each traffic observation point can be regarded as an edge node with the ability to process and perform preliminary analysis of data. By directly preprocessing and performing preliminary analysis of traffic flow data on these edge nodes, the burden on the central server can be greatly reduced and the response speed of the entire system can be improved.

[0004] However, existing edge computing-based traffic flow analysis methods still have some shortcomings. On the one hand, these methods often lack an effective verification mechanism for edge node data processing results, which may lead to invalid or erroneous data being misused, affecting the accuracy of the analysis results. On the other hand, there is a lack of effective storage and management solutions for the large amount of data generated by edge computing, making it difficult to support long-term data analysis and mining. Summary of the invention

[0005] In view of the above-mentioned problems, in combination with the first aspect of the present application, an embodiment of the present application provides a road traffic analysis method based on edge computing, the method comprising:

[0006] Obtaining road traffic data uploaded in real time in a traffic monitoring network; the road traffic data includes X traffic observation points, and the X traffic observation points all record data under the triggering of a first traffic flow update activity; the traffic observation points include the road section traffic flow information corresponding to the traffic monitoring network; X is an integer greater than 1;

[0007] Under the triggering of the first traffic flow update activity, collecting and recording the data of the X traffic flow observation points; and,

[0008] Obtain edge computing data corresponding to the first traffic flow update activity generated by edge computing processing performed by an edge computing device deployed at each traffic flow observation point on the data of the traffic flow observation point, where the edge computing data corresponding to the first traffic flow update activity represents a real-time traffic status update description of the X traffic flow observation points;

[0009] If the edge computing data corresponding to the first traffic flow update activity represents that the real-time traffic status update descriptions of the X traffic observation points are all valid status descriptions, then a data cloud storage configuration related to the first traffic flow update activity is performed in the traffic data management center corresponding to the traffic monitoring network; the data cloud storage configuration related to the first traffic flow update activity represents the edge computing records of the traffic flow data collected by the X traffic observation points stored in the traffic data management center.

[0010] For example, in a possible implementation of the first aspect, the method further includes:

[0011] Acquire a target road traffic perception flow described by a set type based on edge computing records of traffic flow data within a set time period in the traffic data management center in real-time traffic status update description;

[0012] The target road traffic perception flow is loaded into the traffic congestion knowledge point description network, and target traffic congestion knowledge point data is obtained for each traffic congestion knowledge point, wherein the target traffic congestion knowledge point data indicates whether the traffic congestion characteristics of the target road traffic perception flow conform to the traffic congestion knowledge point;

[0013] For example, in a possible implementation of the first aspect, the traffic congestion knowledge point description network includes a semantic encoding unit, a traffic congestion semantic enhancement unit, and a fully connected mapping unit;

[0014] The step of loading the target road traffic perception flow into the traffic congestion knowledge point description network and obtaining target traffic congestion knowledge point data for each traffic congestion knowledge point includes:

[0015] A Gaussian filtering algorithm is used to remove noise in the target road traffic perception stream to generate a noise-cleaned perception stream, and the contrast of the noise-cleaned perception stream is adjusted based on a preset contrast adjustment strategy to highlight key elements in the traffic scene corresponding to the noise-cleaned perception stream to generate a preprocessed perception stream;

[0016] A multi-layer convolutional layer in a convolutional neural network is used to capture different features in the preprocessed perception stream, and an activation function is used to perform nonlinear transformation on the captured different features, and then a maximum pooling operation is used to downsample the output of the last convolutional layer, and the output of the last convolutional layer is used as a preliminary feature representation, and a global average pooling operation is performed on the preliminary feature representation to obtain the average value of each feature channel, and the weight of each feature channel is calculated using an attention mechanism according to the average value of each feature channel, and the weight of each feature channel is normalized using a softmax function to obtain the attention weight of each feature channel;

[0017] fusing the preliminary feature representation with the attention weight of each corresponding feature channel to obtain a weighted feature representation, and fusing the weighted feature representation to obtain a traffic perception attention feature;

[0018] Loading the traffic perception attention feature into the semantic coding unit to generate a semantic coding result, using the semantic coding result from the semantic coding unit as an input of the traffic congestion semantic enhancement unit to generate an initial individual population, wherein each individual in the initial individual population is a transformed representation of the input semantic coding result;

[0019] Calculating a feature correlation parameter between the transformed representation feature generated by each individual and the known semantic feature of traffic congestion and a feature uniqueness parameter of each individual, wherein the feature uniqueness is used to characterize the degree of difference from the transformed representation features generated by other individuals in the initial individual population;

[0020] The feature correlation parameter and the feature uniqueness parameter are weightedly fused to obtain the fitness value of each individual;

[0021] A plurality of individuals are randomly selected from the initial individual population for competition, the individuals in the initial individual population are sorted according to the fitness value of each individual, a target individual with a higher sorting is selected as a parent individual based on a set ratio, and for the selected parent individual, a uniform exchange is performed at a corresponding vector element position with a first set probability to generate a new child individual, and a vector element of the child individual is generated by linearly combining the vector elements of the parent individual;

[0022] Adding random noise to the vector elements of the offspring individuals according to a second set probability to achieve mutation, wherein the second set probability is dynamically adjusted according to the fitness value of the offspring individuals, and the offspring individuals with low fitness values ​​have a higher second set probability;

[0023] Decode the offspring individuals generated by genetic operations and convert them into processable feature representations;

[0024] The decoded processable feature representation is decomposed using principal component analysis to extract the main component features and independent component features. The decomposed main component features and independent component features are screened and filtered according to the pre-set traffic congestion semantic rules to remove some component features that are irrelevant or weakly related to the traffic congestion semantics. The screened component features are reorganized according to the target strategy, and the key components in the screened component features are enhanced through nonlinear transformation to obtain the final traffic congestion semantic enhancement features.

[0025] The traffic congestion semantic enhancement feature is loaded into the fully connected mapping unit, and the target traffic congestion knowledge point data is generated for each traffic congestion knowledge point.

[0026] On the other hand, an embodiment of the present application also provides a road traffic analysis system based on edge computing, including a processor and a machine-readable storage medium, wherein the machine-readable storage medium is connected to the processor, the machine-readable storage medium is used to store programs, instructions or codes, and the processor is used to execute the programs, instructions or codes in the machine-readable storage medium to implement the above method.

[0027] Based on the above aspects, the embodiment of the present application significantly improves the real-time and accuracy of traffic status monitoring by acquiring and processing road traffic data in the traffic monitoring network in real time. The method uses edge computing technology to quickly process data at a location close to the data source, effectively reducing data transmission delays and improving system response speed. At the same time, by uniformly collecting and recording data from multiple traffic observation points under the triggering of the first traffic flow update activity, the synchronization and integrity of the data are ensured, providing a reliable data basis for subsequent traffic status analysis. Further, by edge computing the collected data, edge computing data representing the real-time traffic status update description is generated. This instant feedback mechanism helps to quickly respond to traffic changes and take necessary traffic management measures, such as adjusting the timing of traffic lights and guiding vehicle diversion, thereby effectively alleviating traffic congestion and improving road traffic efficiency. In addition, through the data cloud storage configuration, the long-term preservation and management of effective edge computing records are realized, which not only facilitates the traffic management department to analyze historical data and formulate more scientific traffic planning strategies, but also provides a wealth of research data resources for traffic scientific researchers, promoting technological progress and innovative development in the field of transportation. As a result, the real-time, accuracy and response speed of traffic monitoring can be effectively improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figure 1 It is a schematic diagram of the execution flow of the road traffic analysis method based on edge computing provided in an embodiment of the present application.

[0029] Figure 2This is a schematic diagram of the hardware architecture of the road traffic analysis system based on edge computing provided in an embodiment of the present application. DETAILED DESCRIPTION

[0030] The present application will be described in detail below with reference to the accompanying drawings. Figure 1 This is a flow chart of a road traffic analysis method based on edge computing provided by an embodiment of the present application. The road traffic analysis method based on edge computing is introduced in detail below.

[0031] Step S110, obtaining the road traffic data uploaded in real time in the traffic monitoring network. The road traffic data includes X traffic observation points, and the X traffic observation points all record data under the triggering of the first traffic flow update activity. The traffic observation points contain the road section traffic flow information corresponding to the traffic monitoring network. X is an integer greater than 1.

[0032] In detail, the traffic monitoring network is a comprehensive network composed of multiple traffic monitoring devices and systems, which is used to monitor and manage urban traffic conditions in real time. This network usually includes multiple sensors and devices such as cameras, induction coils, radars, license plate recognition systems, etc., which are distributed in key locations such as the main roads, intersections, and highway entrances and exits in the city. These devices can collect multiple information such as traffic flow, vehicle speed, vehicle model, vehicle driving trajectory, etc. in real time, and transmit this information to the central server or traffic management center through wireless communication or wired network for processing and analysis.

[0033] The road flow data refers to the data collected through the traffic monitoring network about the flow of vehicles on the road. These data include the number of vehicles, vehicle types, speeds, driving directions, driving times and other information. Road flow data is an important basis for traffic management and planning. By analyzing these data, we can understand the traffic conditions, congestion, traffic capacity and other information of the road, and provide support for traffic decision-making.

[0034] The traffic observation point refers to a specific location in the traffic monitoring network used to collect road traffic data. These observation points are usually set at key locations on the road, such as intersections of major roads, entrances and exits of highways, sections of urban expressways, etc. Each observation point is equipped with corresponding sensors and data acquisition equipment, which can obtain real-time vehicle traffic information passing through the observation point, and send this information to the central server or traffic management center through the data transmission system for processing and analysis.

[0035] The first traffic flow update activity refers to a traffic flow data collection and update operation performed according to a predetermined plan or trigger condition. This activity usually occurs in a specific time period, such as the morning and evening rush hours of each day, or a time period temporarily set according to traffic management needs. In the first traffic flow update activity, all set traffic observation points will start recording data at the same time, and the collected data will be transmitted to the central server or traffic management center for processing and analysis. The main purpose of this activity is to obtain the latest road traffic data to reflect changes in current traffic conditions.

[0036] That is, this embodiment assumes that there is a traffic monitoring network in a city, in which the server is responsible for processing and analyzing road flow data. The city is divided into multiple areas, and each area is set with a certain number of flow observation points.

[0037] At a certain moment, the first traffic flow update activity is triggered. For example, this may be a specific time period every day, such as the morning rush hour from 7 to 9 a.m. During this period, 100 traffic observation points (X=100) distributed throughout the city start recording data.

[0038] These traffic observation points may include key locations such as intersections of major roads, entrances and exits of highways, sections of urban expressways, etc. Each traffic observation point is equipped with sensors and data collection equipment, which can obtain real-time information such as the number of vehicles, speed, and vehicle model.

[0039] Take an intersection in the city center as an example. Its traffic observation point detects the number of vehicles passing through the intersection within a certain period of time, the traffic flow ratio in different directions, and the average speed of vehicles through induction coils or cameras buried in the road. At the same time, at the entrances and exits of highways in the suburbs of the city, traffic observation points use electronic toll collection systems and license plate recognition technology to record the number of vehicles entering and leaving the highway, vehicle type, and time information.

[0040] The server obtains the road traffic data uploaded by these traffic observation points in real time through the communication connection established with them. These road traffic data are transmitted in a certain format and protocol, and the server receives and performs preliminary analysis and sorting on them to prepare for subsequent processing.

[0041] Step S120: collecting and recording data of the X traffic flow observation points under the triggering of the first traffic flow update activity.

[0042] In this embodiment, after receiving the trigger signal of the first traffic flow update activity, the server starts to collect and record data from 100 traffic observation points. For each traffic observation point, the server receives the uploaded data at a predetermined time interval (for example, every 1 minute). When the server receives data from the traffic observation point at the intersection in the city center, it records information including the number of vehicles, vehicle speed, and vehicle flow ratio in a dedicated data storage area, and marks the corresponding observation point number and timestamp. At the same time, the server also accurately records the data uploaded by the traffic observation points at the entrances and exits of the highway, including detailed information such as the entry and exit time of the vehicle, vehicle type classification, and toll amount.

[0043] In the process of collecting data, the server needs to ensure the integrity and accuracy of the data. If the data of a certain observation point is lost or erroneous during the transmission process, the server will send a request to the observation point to re-upload the data to ensure the quality of the data.

[0044] The server continuously receives and records data from 100 traffic observation points, forming a huge real-time traffic flow data set, which will provide a basis for subsequent analysis and processing.

[0045] Step S130, obtaining edge computing data corresponding to the first traffic flow update activity generated after edge computing processing is performed on the data of the traffic observation point by the edge computing device deployed at each traffic observation point, wherein the edge computing data corresponding to the first traffic flow update activity represents a real-time traffic status update description of the X traffic observation points.

[0046] In this embodiment, after completing the collection of data from 100 traffic observation points, the server immediately starts the edge computing process through the edge computing devices deployed at each traffic observation point. The purpose of edge computing is to quickly perform preliminary analysis and processing on the collected data, extract valuable information, and reflect the updated traffic status in real time. The edge computing device first pre-processes the data from each observation point, such as removing outliers and filling missing values. Then, through a series of algorithms and models, the traffic flow indicators of each observation point, such as traffic flow density, average vehicle speed, congestion index, etc., are calculated.

[0047] Taking a busy main road observation point as an example, the edge computing device calculates the traffic flow density of the road section based on the collected vehicle number and speed data. If the number of vehicles is large and the speed is slow within a certain period of time, the congestion index will increase accordingly, indicating that the road section is congested.

[0048] For observation points at the entrances and exits of highways, edge computing devices calculate the traffic efficiency of the entrances and exits based on the time and number of vehicles entering and exiting. If the waiting time of vehicles in queue is long and the traffic efficiency is low, this will also be reflected in the edge computing data.

[0049] Based on the calculation results of 100 observation points, the edge computing device generates edge computing data corresponding to the first traffic flow update activity. These edge computing data describe the real-time traffic status update of each observation point in a structured manner, providing a key basis for subsequent decision-making and storage, and are uploaded to the server.

[0050] Step S140: If the edge computing data corresponding to the first traffic flow update activity indicates that the real-time traffic status update descriptions of the X traffic observation points are all valid status descriptions, then a data cloud storage configuration related to the first traffic flow update activity is performed in the traffic data management center corresponding to the traffic monitoring network. The data cloud storage configuration related to the first traffic flow update activity indicates that the edge computing records of the traffic flow data collected by the X traffic observation points are stored in the traffic data management center.

[0051] In this embodiment, after completing edge computing and generating relevant data, the server will evaluate the validity of the data.

[0052] If the real-time traffic status update descriptions of all 100 traffic observation points are considered valid, the server will start the data cloud storage configuration process.

[0053] For example, the server confirms that the data from all observation points accurately reflects the actual traffic conditions without obvious errors or anomalies. In this case, the server establishes a connection with the traffic data management center and securely stores the collected traffic flow data from 100 observation points and the related indicators and description information generated by edge computing to the cloud.

[0054] During the storage process, the server will organize and classify the data according to certain rules and formats to facilitate subsequent query and analysis. For example, the observation point data of different regions and different types of roads are stored in different data tables, and indexes are established to improve query efficiency. At the same time, the server will also record metadata information such as the storage time, source, and processing process of the data to ensure the traceability and integrity of the data. By storing these data in the cloud of the traffic data management center, not only can the long-term preservation and backup of the data be achieved, but it can also facilitate traffic management departments, research institutions and other parties to access and use these data at any time, providing strong support for the planning, management and optimization of urban transportation.

[0055] Assume that in the above process, the server finds that the data of several observation points are abnormal, such as the number of vehicles at a certain observation point is obviously deviated from the normal range, or the speed data has unreasonable fluctuations. In this case, the server will consider the real-time traffic status update description of these observation points as invalid status description, and will not store the relevant data in the cloud, but will conduct further investigation and processing to ensure the accuracy and reliability of the data.

[0056] In a possible implementation, the method further includes: if the edge computing data corresponding to the first traffic flow update activity indicates that there is a traffic observation point with an invalid state description among the X traffic observation points, then revocation of the data cloud storage related to the first traffic flow update activity is performed in the traffic data management center corresponding to the traffic monitoring network. The revocation of the data cloud storage related to the first traffic flow update activity indicates that the edge computing records of the traffic flow data collected by the X traffic observation points are revoked in the traffic data management center.

[0057] Among them, after the edge computing records of the traffic flow data collected by the X traffic observation points are configured to the traffic data management center by cloud storage, the traffic management service node can retrieve the traffic flow observation data related to the X traffic observation points from the traffic data management center. The edge computing records of the traffic flow data collected by the X traffic observation points include at least one of the following: a new record of adding the traffic flow observation data related to the X traffic observation points to the traffic data management center. A revocation record of revoking the traffic flow observation data related to the X traffic observation points from the traffic data management center. An adjustment record of adjusting the traffic flow observation data related to the X traffic observation points. And a search record for the traffic flow observation data related to the X traffic observation points in the traffic data management center.

[0058] In this embodiment, in the aforementioned urban traffic monitoring network, the server has completed the edge computing of the data of 100 traffic observation points in the first traffic flow update activity. However, when evaluating these edge computing data, the server finds that the real-time traffic status update description of 5 traffic observation points is invalid.

[0059] Taking an observation point at an intersection in the suburbs of a city as an example, the server found that the number of vehicles uploaded by this observation point was significantly lower than the historical level during the same period, and the speed data was abnormally high, which was seriously inconsistent with the usual traffic conditions at the intersection. After further analysis and judgment, the server determined that the data at this observation point was invalid.

[0060] Due to the existence of such invalid data observation points, the server decides to revoke the data cloud storage related to the first traffic flow update activity in the traffic data management center corresponding to the traffic monitoring network. This means that the server will revoke the edge computing records of the traffic flow data collected by these 100 traffic observation points that may have been stored in the cloud.

[0061] Specifically, if some data has been added to the traffic data management center before cloud storage is configured, such as normal and valid vehicle quantity, vehicle speed and traffic flow ratio data from another intersection observation point in the city center, the server will cancel these new records.

[0062] At the same time, if there are previous records of adjustments to certain data, such as corrections to the vehicle type classification at a highway entrance or exit observation point, the server will also revoke such adjustment records and restore them to the state before the adjustment.

[0063] In addition, if the traffic management service node has previously searched for traffic observation data related to these 100 traffic observation points in the traffic data management center and formed a search record, the server will also revoke these search records to ensure the consistency and accuracy of the data.

[0064] For example, when the traffic management staff queried the relevant data, they might have been able to obtain the preliminary analysis results of the 100 traffic observation points. However, due to the invalid data of some observation points, the cloud storage was revoked. When they queried again, they would not be able to obtain the data related to the first traffic flow update activity.

[0065] Through such data cloud storage revocation operations, the server can ensure that the data in the traffic data management center is always accurate and valid, providing reliable support for subsequent traffic planning, management and decision-making, and avoiding erroneous analysis and decision-making due to the existence of some invalid data.

[0066] For example, if the city traffic planning department originally planned to plan new road construction based on this traffic data, but it was revoked because some of the data was invalid, they would not make wrong planning decisions based on inaccurate data, but wait for the next valid and complete traffic flow update activity data to develop a more reasonable and scientific traffic planning scheme.

[0067] For example, the traffic light control system originally planned to adjust the duration of traffic lights at certain intersections based on the traffic data. However, because some data was invalid, the relevant adjustment plan would be suspended until accurate and valid data was obtained, and then optimized to ensure smooth and safe traffic.

[0068] Therefore, by timely discovering and processing invalid traffic observation point data and performing data cloud storage revocation operations, the quality and reliability of traffic data are effectively maintained, providing a strong guarantee for the efficient operation of urban transportation.

[0069] Based on the above steps, the embodiment of the present application significantly improves the real-time and accuracy of traffic status monitoring by acquiring and processing road traffic data in the traffic monitoring network in real time. The method uses edge computing technology to quickly process data at a location close to the data source, effectively reducing data transmission delays and improving system response speed. At the same time, by uniformly collecting and recording data from multiple traffic observation points under the triggering of the first traffic flow update activity, the synchronization and integrity of the data are ensured, providing a reliable data basis for subsequent traffic status analysis. Further, by edge computing the collected data, edge computing data representing the real-time traffic status update description is generated. This instant feedback mechanism helps to quickly respond to traffic changes and take necessary traffic management measures, such as adjusting the timing of traffic lights and guiding vehicle diversion, thereby effectively alleviating traffic congestion and improving road traffic efficiency. In addition, through the data cloud storage configuration, the long-term preservation and management of effective edge computing records are realized, which not only facilitates the traffic management department to analyze historical data and formulate more scientific traffic planning strategies, but also provides a wealth of research data resources for traffic scientific researchers, promoting technological progress and innovative development in the field of transportation. As a result, the real-time, accuracy and response speed of traffic monitoring can be effectively improved.

[0070] In a possible implementation, the method further includes:

[0071] Step A110, constructing data access rights for the road traffic data. According to the data access rights, executing a permission protection instruction on the traffic observation data related to the X traffic observation points in the data collection process, the traffic observation data on which the permission protection instruction is executed is in a permission protection state.

[0072] Step A120: When the traffic observation data related to the X traffic observation points are in an authority protection state, the data of the X traffic observation points are collected and recorded under the triggering of the first traffic flow update activity.

[0073] In this embodiment, in the urban traffic monitoring network set in the above embodiments, the server undertakes the important task of constructing data access rights for road traffic data.

[0074] First, the server obtains the traffic data mapping bitmap of the traffic data management center corresponding to the traffic monitoring network. This bitmap records in detail the data index information of at least one road traffic data including the current real-time uploaded road traffic data. From this huge traffic data mapping bitmap, the server accurately determines the data index information of the real-time uploaded road traffic data.

[0075] Based on this clear data index information, the server configures strict data access permissions for the real-time uploaded road traffic data. For example, for traffic observation data at an important intersection in the city center, the server sets that only senior administrators of a specific traffic management department and authorized data analysis teams can access and process this data. For traffic observation data at the entrances and exits of suburban highways, only specific personnel from the traffic planning department may be allowed to access it during a specific time period.

[0076] When the traffic observation data related to the X traffic observation points are in this strict authority protection state, the first traffic flow update activity is triggered. The server starts to collect and record the data of the X traffic observation points according to the predetermined rules and procedures.

[0077] Take 10 traffic observation points located in the city's commercial district as an example. These observation points cover key sections around main streets and shopping malls. In the permission protection state, the server receives data from these observation points. For example, one of the observation points uploads detailed information such as the number of vehicles, vehicle models, and average speed passing through the section during a specific time period. After confirming that the access rights are correct, the server accurately records this data.

[0078] For the five traffic observation points on the main bridges in the city, they uploaded data on traffic flow, vehicle load, etc. Due to the pre-set permission protection, the server will strictly follow the permission rules when processing this data to ensure that only legally authorized personnel or systems can obtain and use this data.

[0079] Throughout the data collection process, the server ensures that any unauthorized access requests are denied. Even in emergency situations, such as emergency response to traffic emergencies, the server will first verify the access rights of the relevant personnel or system before deciding whether to provide the data.

[0080] For example, in a traffic accident that caused a local traffic paralysis, the emergency rescue department needs to obtain real-time traffic data for a specific road section. After receiving the request, the server quickly verifies its authority. After confirming that it has legitimate emergency access rights, the server provides the relevant data to the emergency rescue department to assist them in formulating effective rescue and traffic diversion plans.

[0081] At the same time, the server will also keep detailed records of data access and use. No matter who accesses the data, when, and for what purpose, it will be recorded by the server. This not only helps with subsequent auditing and tracing, but also further strengthens data security and compliance management.

[0082] Therefore, by constructing data access permissions for road flow data, the flow observation data related to X flow observation points are effectively collected and recorded under the premise of ensuring data security and privacy, providing a solid data foundation for the management and optimization of urban traffic.

[0083] In a possible implementation, step A110 includes:

[0084] Step A111, obtaining a flow data mapping bitmap of a traffic data management center corresponding to the traffic monitoring network. The flow data mapping bitmap is used to record data index information of at least one road flow data, wherein the at least one road flow data includes the road flow data uploaded in real time.

[0085] Step A112: determining data index information of the road traffic data uploaded in real time from the traffic data mapping bitmap.

[0086] Step A113: configuring the data access permission for the road traffic data uploaded in real time according to the data index information of the road traffic data uploaded in real time.

[0087] In the city's traffic monitoring network, first, the server starts the process of obtaining the traffic data mapping bitmap of the traffic data management center corresponding to the traffic monitoring network. This traffic data mapping bitmap is like a huge information warehouse index, which densely records the data index information of at least one road traffic data. The server successfully obtains this crucial bitmap through a stable network connection and a specific interface protocol.

[0088] After obtaining the bitmap, the server begins to carefully filter and search in this massive amount of data index information. For example, the server needs to determine the data index information of the road traffic data uploaded in real time from the main roads of the city from a large amount of information. Through a series of complex algorithms and matching logic, the server quickly and accurately finds the corresponding specific identification and code, thereby successfully determining the index information of these key data.

[0089] Next, based on these precisely determined data index information, the server sets out to configure the corresponding data access permissions for the real-time uploaded road traffic data. For example, for the traffic data of a main road with extremely high traffic volume in the city's core business district, the server sets that only the core members of the city's traffic planning department and senior traffic management officials can obtain complete data and conduct in-depth analysis for the formulation of major traffic plans and decisions. For the traffic data of some secondary roads or roads near residential areas, grassroots traffic management personnel may be allowed to review it during specific working hours so that they can adjust local traffic diversion measures in a timely manner.

[0090] Taking the traffic data of urban expressways as an example, the server configures read-only permissions for relevant research institutions so that they can study and optimize traffic flow models based on this data, but they are not allowed to modify the data. For some sensitive areas, such as traffic data on roads near government agencies, the server sets extremely strict access rights, which can only be accessed and used under certain security levels and authorization conditions.

[0091] When configuring permissions for traffic data on a city's import and export roads, the server may allow logistics companies to obtain some data during a specific time period so that they can optimize the routes and schedules of cargo transportation, but at the same time will strictly limit and monitor the scope and method of data use.

[0092] In addition, the server will dynamically adjust the permission configuration according to different times and scenarios. For example, during holidays or major events, the server will temporarily expand the access and operation permissions of relevant traffic management departments and emergency service agencies for road traffic data leading to tourist attractions and event venues to ensure smooth and safe traffic.

[0093] Throughout the entire process, the server always maintains a high degree of accuracy and security. It continuously verifies and optimizes the permission configuration to ensure that each piece of road traffic data can be reasonably used within the appropriate scope of authority, providing strong support for the city's traffic management and development.

[0094] Based on this design, a scientific, reasonable, safe and reliable data access permission system is established for road traffic data in the urban traffic monitoring network.

[0095] In a possible implementation, step A110 includes:

[0096] Step A114, obtaining a flow data mapping bitmap of a traffic data management center corresponding to the traffic monitoring network. The flow data mapping bitmap is used to record data index information of at least one road flow data, wherein the at least one road flow data includes the road flow data uploaded in real time.

[0097] Step A115, configuring the data access permission for the traffic data mapping bitmap, wherein the road traffic data corresponding to each data index information in the traffic data mapping bitmap is configured with the data access permission.

[0098] In this embodiment, the server first obtains the flow data mapping bitmap of the traffic data management center corresponding to the traffic monitoring network. This flow data mapping bitmap is like a super large database directory, which densely lists the data index information of at least one road flow data. The server establishes a close connection with the traffic data management center through an efficient and stable internal communication line. After a series of identity authentication and data request processes, this crucial flow data mapping bitmap is successfully captured.

[0099] This traffic data mapping bitmap contains index information of road traffic data from all corners of the city. Next, the server starts to configure data access permissions for the entire traffic data mapping bitmap. This means that the road traffic data corresponding to each data index information in the bitmap will be assigned specific permission rules.

[0100] For the main transportation hubs in the city, such as road traffic data near high-speed rail stations, the server has set a high level of access rights. Only senior leaders of the transportation department and specific emergency command teams can obtain real-time and detailed data so that they can make decisions quickly during major holidays or emergencies. For example, on the eve of the Spring Festival holiday, in order to cope with the huge passenger flow that may occur, relevant personnel can use their permissions to obtain accurate traffic data in the area, so as to reasonably allocate transportation resources.

[0101] The server gives the school security department and local traffic police access to traffic data on roads around the school during the school commute period. They can arrange traffic diversion personnel based on this data to ensure students' travel safety. For example, from 7 to 8 a.m. and 4 to 5 p.m. every day, the school security department and traffic police can check the data in time, understand the changes in traffic flow, and make preparations in advance.

[0102] The traffic data of the city's expressway is of great significance to the city's overall traffic planning. The server grants specific read-only permissions to the city planning department and traffic research institutions, allowing them to analyze and study the data within a specific time period, but prohibiting them from modifying the original data. For example, during working hours from Monday to Friday, researchers from the city planning department can log in to the system and obtain this data to optimize the city's traffic planning model.

[0103] For some areas where road construction is ongoing, the server strictly restricts access rights. Only the person in charge of the construction project and the supervisor designated by the traffic management department can obtain relevant traffic data to ensure that traffic adjustment measures during the construction period can be implemented scientifically and effectively. For example, during a one-month renovation project on a certain road, relevant personnel can adjust the construction progress and the setting of temporary traffic signs based on real-time traffic data.

[0104] For some roads involving national security or sensitive areas, the server sets the most stringent access rights. Only after multiple layers of approval and specific security verification processes can a very small number of authorized personnel obtain limited data. For example, on roads near a military control area, only under special circumstances and after strict security review can relevant departments obtain necessary traffic data to ensure regional security and normal traffic operation.

[0105] By configuring fine-grained and appropriate data access permissions for each data index information in the traffic data mapping bitmap, the security, accuracy and rational use of road traffic data are ensured, providing solid and reliable support for urban traffic management and development.

[0106] In a possible implementation, under the condition that the data cloud storage configuration related to the first traffic flow update activity is performed in the traffic data management center corresponding to the traffic monitoring network, or under the condition that the data cloud storage related to the first traffic flow update activity is revoked in the traffic data management center corresponding to the traffic monitoring network, the method further includes: releasing the permission protection instruction of the data access right for the traffic observation data related to the X traffic observation points. After the permission protection instruction is released, the traffic observation data related to the X traffic observation points is configured in a modifiable state.

[0107] In this embodiment, in the city's traffic monitoring network, the server will perform corresponding operations according to different situations when processing data related to the first traffic flow update activity.

[0108] Assume that the server has determined that the conditions for performing data cloud storage configuration related to the first traffic flow update activity in the traffic data management center corresponding to the traffic monitoring network are met. For example, after evaluating the edge computing data from 100 traffic observation points in the first traffic flow update activity, the server confirms that the real-time traffic status update descriptions of all data are valid status descriptions. Then, the server starts to perform the data cloud storage configuration operation, and securely stores the edge computing records of the traffic flow data collected by these traffic observation points in the cloud of the traffic data management center.

[0109] After completing the data cloud storage configuration, the server, in accordance with the established rules, lifted the data access rights and permission protection instructions for the traffic observation data related to these 100 traffic observation points. Take one of the traffic observation points located at the intersection of the main road and the secondary road in the city as an example. Previously, its data was in a strict permission protection state, and only specific senior traffic management accounts could access it. Now, the server has lifted the permission protection instructions, and the data of this observation point is configured to be modifiable.

[0110] This means that authorized traffic data analysts can now modify and adjust the data. For example, if they find that the traffic flow data recorded at the observation point for a certain period of time has a slight deviation, it may be caused by a short-term failure of the sensor. When the data is modifiable, they can correct this part of the data to ensure the accuracy and reliability of the data.

[0111] Assume that the server determines that it is necessary to perform a data cloud storage revocation operation related to the first traffic flow update activity in the traffic data management center. For example, the server finds that the real-time traffic status update descriptions of 10 observation points out of the 100 traffic flow observation points are invalid status descriptions, which may be due to data transmission interruption or serious failure of the observation equipment. In this case, the server performs a data cloud storage revocation operation to withdraw the relevant data that may have been stored in the cloud.

[0112] After the data cloud storage is revoked, the server also removes the permission protection instructions for the data access rights that were previously set. Take one of the highway entrance and exit observation points that was determined to be invalid as an example. Previously, its data access was strictly restricted. Now the server removes the protection instructions and configures its data to be modifiable. Relevant technicians can further investigate and analyze the data to find out the root cause of the invalid data, and repair or replace the observation equipment to prepare for the next traffic flow update activity.

[0113] Throughout the entire process, the server strictly adheres to the principles of security and compliance. The instructions for removing permission protection and configuring data to be editable are recorded in detail in the server's operation log, including the time of execution, the identity information of the operator, and the specific content of the modification. This not only facilitates subsequent auditing and tracing, but also ensures the transparency and fairness of the data processing process.

[0114] For example, when the urban traffic planning department formulates a new traffic planning scheme, it needs to conduct in-depth analysis and adjustment of the data of the first traffic flow update activity that was previously stored. Since the server has lifted the permission protection instructions and configured the data to be modifiable, the professionals of the planning department can flexibly process the data according to actual needs, such as recalculating the traffic flow indicators of certain key sections, or adjusting the statistical time period of the data to obtain more targeted information.

[0115] For example, when responding to an unexpected traffic incident, the traffic management department discovered that there was a problem with the traffic data of an important intersection in the first traffic flow update activity that was previously stored. After the server removed the permission protection and configured the modifiable state, they were able to quickly correct and improve the data, thereby more accurately assessing the traffic conditions and formulating effective emergency measures, such as temporarily adjusting the duration of traffic lights or dispatching additional traffic police to conduct on-site traffic diversion.

[0116] By removing protection instructions for data access permissions under appropriate conditions and configuring relevant traffic observation data to be modifiable, the server provides strong support for the scientific management and optimization of urban traffic, ensuring that traffic data can better serve actual traffic decision-making and planning.

[0117] In a possible implementation, the road traffic data further includes Y traffic observation points, the recording timing of the Y traffic observation points is located after the X traffic observation points, and the Y traffic observation points all record data under the triggering of the second traffic flow update activity. Y is an integer greater than 1. The method further includes:

[0118] Step B110, under the triggering of the second traffic flow update activity, recording the traffic flow data of the Y traffic flow observation periods.

[0119] Step B120: Perform edge computing processing on the data collection process corresponding to the second traffic flow update activity to generate edge computing data corresponding to the second traffic flow update activity. The edge computing data corresponding to the second traffic flow update activity represents the real-time traffic status update description of the Y traffic observation points.

[0120] Step B130: If the edge computing data corresponding to the second traffic flow update activity represents that the real-time traffic status update descriptions of the Y traffic observation points are all valid status descriptions, then a data cloud storage configuration related to the second traffic flow update activity is performed in the traffic data management center corresponding to the traffic monitoring network. The data cloud storage configuration related to the second traffic flow update activity represents the edge computing records of the traffic flow data of the Y traffic observation periods stored in the traffic data management center.

[0121] The trigger requirement for terminating the recording of the road traffic data is that the traffic flow data of the last traffic observation point in the road traffic data is completely recorded.

[0122] In this embodiment, in the aforementioned urban traffic monitoring network, after the relevant processing of the first traffic flow update activity is completed, the second traffic flow update activity is ushered in.

[0123] Assume that the 100 traffic observation points (X=100) involved in the first traffic flow update activity are for data collection during the morning rush hour (7:00 to 9:00), and this time the Y traffic observation points are distributed in different areas of the city to monitor the midday traffic flow from 12:00 noon to 2:00 pm. For example, Y is 80.

[0124] When the second traffic flow update activity is triggered, the server starts recording data from the 80 traffic flow observation points. For example, one of the traffic flow observation points located near the city's commercial center records the number, type, and speed of vehicles passing through the road section in real time through its high-definition cameras and sensing equipment. The server receives and records data from this observation point at a predetermined frequency (for example, every 30 seconds).

[0125] At the same time, in another area of ​​the city, a traffic observation point located near a school is also uploading data. The server also accurately records the traffic flow data of this section of road when students are out of school, including the changing trends of traffic flow and the staying time of vehicles.

[0126] The server continuously receives and records data from these 80 traffic observation points to form a noon traffic flow dataset.

[0127] Next, the server performs edge computing on the data collection process corresponding to the second traffic flow update activity. It first preprocesses the data to remove outliers and supplement possible missing values.

[0128] For an observation point located on a main road in a city, the server calculates the traffic density and average travel time of the road section based on the collected vehicle quantity and speed data. If there are a large number of vehicles and a long travel time in a certain period of time, the server will determine that the road section may be congested and make corresponding annotations in the edge computing data.

[0129] For an observation point located near an urban industrial park, the server analyzes the traffic congestion and transportation demand in the area by combining vehicle types (such as the ratio of trucks to buses) and changes in traffic flow.

[0130] Combining the calculation results of these 80 traffic observation points, the server generates edge computing data corresponding to the second traffic flow update activity. These data clearly depict the real-time traffic status update of each observation point.

[0131] The server evaluated the edge computing data and found that the real-time traffic status update descriptions of the 80 traffic observation points were all valid status descriptions. For example, the data of all observation points conformed to the usual traffic patterns of their locations and time periods, with no obvious anomalies or errors.

[0132] Since the data is valid, the server decides to configure the cloud storage of data related to the second traffic flow update activity in the traffic data management center corresponding to the traffic monitoring network. The server establishes a secure connection with the data management center and stores the traffic flow data collected by the 80 traffic observation points at noon and the related indicators and description information generated by edge computing in an orderly manner in the cloud.

[0133] During the storage process, the server classifies and organizes the data according to established rules and formats. For example, the observation point data in the commercial center area is classified into one category, and the observation point data around the school is classified into another category, so that subsequent query and analysis can be more efficient and accurate.

[0134] At the same time, the server will also record detailed metadata information, including data collection time, location information of observation points, data processing process, etc., to ensure the traceability and integrity of the data.

[0135] The entire road traffic data recording process will continue until the traffic flow data of the last traffic observation point is fully recorded, which will trigger the termination of road traffic data recording.

[0136] Assume that one day in the future, the city planning department needs to formulate a new traffic planning scheme. They can obtain multiple traffic flow data including the second traffic flow update activity from the cloud of the traffic data management center, and fully understand the traffic conditions in different time periods and different areas of the city, so as to formulate a more scientific and reasonable traffic plan, such as adding new roads, optimizing bus routes or adjusting traffic light settings.

[0137] For example, when responding to sudden traffic incidents, the traffic management department can quickly retrieve traffic data for relevant time periods and areas, accurately determine the cause and extent of traffic congestion, and take effective diversion measures in a timely manner to ensure smooth and safe traffic.

[0138] Through such an orderly traffic flow data collection, processing, evaluation and storage mechanism, strong data support is provided for urban traffic management and optimization, promoting the efficient operation and sustainable development of urban traffic.

[0139] In a possible implementation, in the data collection process, each of the X traffic observation points is configured with a corresponding data record entry, and the data record entry covers the road section traffic flow information corresponding to the traffic monitoring network contained in the corresponding traffic observation point. Among the X traffic observation points, there is a data record entry corresponding to a traffic observation point associated with a data processing module, and the data processing module covers the prior road section traffic flow information associated with the corresponding traffic observation point, and the prior road section traffic flow information is the road section traffic flow information constructed based on prior knowledge. After the data cloud storage configuration related to the first traffic flow update activity is performed in the traffic data management center corresponding to the traffic monitoring network, the method further includes: activating the data processing module associated with the data record entry.

[0140] In the urban traffic monitoring network, the server is processing the data of the first traffic flow update activity. Assume that in the first traffic flow update activity, there are 100 traffic flow observation points (X=100), and each observation point is configured with a corresponding data record entry.

[0141] Taking one of the traffic observation points located at the intersection of the main road and the secondary road in the city as an example, its data record entries cover in detail the traffic flow information of the road section corresponding to the observation point, such as the number of vehicles passing through in different time periods, vehicle type classification, speed range, etc.

[0142] Among these 100 traffic observation points, the data record entries corresponding to some observation points are associated with data processing modules. For example, an observation point located near a busy business district in the city center has a data record entry associated with a data processing module because the traffic conditions in this area are complex and important. This data processing module covers the priori traffic flow information of the road section associated with the observation point. This priori information is the traffic flow pattern of the road section constructed based on historical data and empirical knowledge.

[0143] After the server completes the data cloud storage configuration related to the first traffic flow update activity in the traffic data management center corresponding to the traffic monitoring network, the server activates the data processing module associated with the data record entry according to the established process.

[0144] The server starts the activation operation for the data processing module associated with the observation point near the downtown business district mentioned above. After receiving the latest traffic data, the data processing module will compare and analyze it with the prior traffic flow information of the road section.

[0145] For example, prior information indicates that between 5pm and 6pm on weekdays, this section of road usually has a large volume of traffic and a low speed. However, the data collected this time shows that the volume of traffic exceeds the previous level during the same period, and the speed is also significantly reduced. The data processing module will perform further calculations and evaluations based on these differences.

[0146] It might calculate the current traffic congestion index and compare it with the historical congestion index of the same period to determine the severity of the current traffic situation. At the same time, the module may also analyze the reasons for this change, such as whether there are new commercial activities nearby that attract more vehicles, or whether there is a decrease in capacity due to road construction.

[0147] In addition, for another data processing module associated with an observation point at the entrance of a city expressway, after being activated, it will analyze the currently collected data in combination with the prior information about the traffic patterns of the entrance at different time periods. If the current traffic growth rate is found to be abnormal, an early warning may be triggered, prompting the traffic management department to pay attention to whether there are potential traffic incidents or abnormal situations.

[0148] By activating these data processing modules associated with data record entries, the server can more deeply explore and analyze the connotation of traffic flow data, providing more valuable decision support for traffic management and planning.

[0149] In a possible implementation, Z data record entries are associated with data processing modules, Z is a positive integer, and Z≤X. The activating the data processing module associated with the data record entry includes: performing distributed parallel processing on the data processing modules associated with the Z data record entries.

[0150] Among them, the data record entry corresponding to the traffic observation point is associated with the data processing module both when the entry is generated and when the data processing is completed.

[0151] Among the data processing modules associated with the Z data record entries, the data processing modules that are processed in distributed parallel include: the data processing modules associated with each data record entry except the data record entry belonging to the target type observation point in the Z data record entries when the entry is generated and when the data processing is completed, and the data processing modules associated with the data record entry belonging to the target type observation point in the Z data record entries at the trigger node of the entry construction. The data processing modules associated with the data record entry belonging to the target type observation point in the Z data record entries at the termination node of the entry construction are activated in the next traffic flow update activity of the first traffic flow update activity.

[0152] In this embodiment, in the traffic monitoring network of the city, it is assumed that in the first traffic flow update activity, there are a total of 100 traffic flow observation points (X=100), of which 50 data record entries are associated with the data processing module (Z=50).

[0153] Among these data record entries associated with the data processing modules, 10 data record entries belong to target type observation points, and the remaining 40 do not belong to target type observation points.

[0154] The server starts to activate the data processing modules associated with these data record entries. First, for the data processing modules associated with the 40 data record entries that do not belong to the target type observation points, the server adopts a distributed parallel processing method.

[0155] For example, one of the observation points located on a road in an ordinary residential area in the city has a data processing module associated with its data record entry when it is generated, which is used to analyze the traffic flow change trend in the area within a specific time period. When the server receives the data from this observation point, it immediately starts the data processing module associated with it for processing. At the same time, the data processing module associated with the data record entry of another observation point located near a small commercial street is also being processed in parallel. It is responsible for calculating the traffic flow density and the average dwell time of vehicles on this road section.

[0156] For the data processing modules associated with the 10 data record entries belonging to the target type observation points, the server associates the corresponding data processing modules when the trigger node of the entry is constructed.

[0157] Take an observation point located near an important transportation hub as an example. When the first traffic flow update activity starts, this observation point starts recording data, which triggers the data processing module associated with its data record entry. However, at this stage, only preliminary processing and analysis are performed on it, such as simple statistics of vehicle flow and calculation of preliminary congestion index.

[0158] The data processing modules associated with the data record entries belonging to the target type observation points at the termination nodes of the entry construction will be activated in the next traffic flow update activity of the first traffic flow update activity.

[0159] For example, when the next traffic flow update activity (assuming it is the afternoon peak hour) arrives, the data processing module associated with the observation point near the previous important transportation hub at the end node of the first traffic flow update activity is activated. This module will combine the data collected this time and in previous times to conduct a more in-depth analysis, such as comparing traffic flow changes in the same period on different dates and evaluating the effectiveness of traffic improvement measures.

[0160] During the entire processing process, the server efficiently coordinates and allocates resources to ensure that each data processing module can process relevant data in a timely and accurate manner. Through this distributed parallel processing method, the efficiency and speed of data processing are greatly improved, and valuable information can be provided for traffic management and decision-making more quickly.

[0161] For another example, an observation point located near a school is a target type observation point in the first traffic flow update activity. At the beginning of the activity, the associated data processing module is triggered to perform preliminary processing and record the vehicle flow and pedestrian flow during the school hours. When the first traffic flow update activity ends, the data processing module associated with the termination node is not activated. When the next traffic flow update activity (such as a special period when the school holds a large-scale event), the module is activated to compare the traffic flow difference between the special period and the normal period, providing a basis for formulating targeted traffic diversion plans.

[0162] For example, an observation point at the entrance and exit of a large factory is not a target type observation point in the first traffic flow update activity. The data processing module associated with its data record entries performs distributed parallel processing when generating entries and completing data processing, and analyzes the impact of factory traffic flow during working hours on surrounding roads in real time, and calculates the average queue length and waiting time of vehicles.

[0163] Through such sophisticated and efficient data processing module activation and processing methods, the server can fully tap the value of traffic flow data and provide strong support for optimizing urban traffic conditions.

[0164] For example, in a possible implementation, the method further includes:

[0165] Step S150, based on the edge computing records of traffic flow data within a set time period in the traffic data management center, obtain the target road traffic perception flow described as a set type of real-time traffic status update description.

[0166] Step S160, loading the target road traffic perception flow into the traffic congestion knowledge point description network, obtaining target traffic congestion knowledge point data for each traffic congestion knowledge point, and the target traffic congestion knowledge point data characterizes whether the traffic congestion characteristics of the target road traffic perception flow conform to the traffic congestion knowledge point.

[0167] In this embodiment, the server first obtains the edge computing records of traffic flow data within a set time period from the traffic data management center. Then, based on the edge computing records, the server selects the target road traffic perception flow whose real-time traffic status update description is described as a set type.

[0168] Suppose one of the preset traffic congestion knowledge points is described as: "From 8 to 9 a.m. on weekdays, the main road in the city center connecting the main commercial district has a traffic volume of more than 1,500 vehicles per hour, an average speed of less than 15 kilometers per hour, and a queue length of more than 500 meters between two adjacent intersections, and this state lasts for at least 30 minutes."

[0169] Next, the server loads the target road traffic perception stream into the traffic congestion knowledge point description network.

[0170] For example, another knowledge point about traffic congestion is described as: "During the weekend from 2:00 p.m. to 5:00 p.m., the roads leading to popular urban attractions have a traffic volume of more than 800 vehicles per hour, an average speed of less than 25 kilometers per hour, and vehicles queue at the entrance of the parking lot near the scenic spot for more than 10 minutes." The server determines whether the target road meets the description of this complex knowledge point based on the specific data in the traffic perception flow, such as time, location, traffic volume, speed, and vehicle queue conditions.

[0171] Suppose there is another knowledge point described as: "During the school rush hours every day when the school is in session, the traffic volume on the roads around the school surges sharply, more than twice that of normal times, the speed is generally less than 20 kilometers per hour, the vehicles on the road section in front of the school are stagnant for more than 15 minutes, and a chain reaction of congestion occurs on the surrounding adjacent roads." The server also determines whether the traffic conditions of the relevant roads match this complex description based on the detailed data collected.

[0172] By comparing and analyzing the target road traffic perception flow with these complex and specific traffic congestion knowledge point descriptions, the server can accurately determine whether the traffic congestion characteristics of the target road meet these preset traffic congestion knowledge points, thereby providing accurate and targeted basis for traffic management and planning.

[0173] For example, in a possible implementation, the traffic congestion knowledge point description network includes a semantic encoding unit, a traffic congestion semantic enhancement unit, and a fully connected mapping unit.

[0174] Step S160 includes:

[0175] Step S161, using a Gaussian filtering algorithm to remove noise in the target road traffic perception flow to generate a noise-cleaned perception flow, and adjusting the contrast of the noise-cleaned perception flow based on a preset contrast adjustment strategy to highlight the key elements in the traffic scene corresponding to the noise-cleaned perception flow, thereby generating a preprocessed perception flow.

[0176] Step S162, using multiple convolutional layers in a convolutional neural network to capture different features in the preprocessed perception stream, and using an activation function to perform nonlinear transformation on the captured different features, and then using a maximum pooling operation to downsample the output of the last convolutional layer, and using the output of the last convolutional layer as a preliminary feature representation, performing a global average pooling operation on the preliminary feature representation to obtain the average value of each feature channel, and using an attention mechanism to calculate the weight of each feature channel according to the average value of each feature channel, and using a softmax function to normalize the weight of each feature channel to obtain the attention weight of each feature channel.

[0177] Step S163, fusing the preliminary feature representation with the attention weight of each corresponding feature channel to obtain a weighted feature representation, and fusing the weighted feature representation to obtain a traffic-aware attention feature.

[0178] Step S164, loading the traffic perception attention feature into the semantic coding unit to generate a semantic coding result, using the semantic coding result from the semantic coding unit as the input of the traffic congestion semantic enhancement unit, generating an initial individual population, wherein each individual in the initial individual population is a transformed representation of the input semantic coding result.

[0179] Step S165, calculating the feature correlation parameter between the transformed representation feature generated by each individual and the known traffic congestion semantic feature and the feature uniqueness parameter of each individual, wherein the feature uniqueness is used to characterize the degree of difference from the transformed representation features generated by other individuals in the initial individual population.

[0180] Step S166, weightedly fuse the feature correlation parameter and the feature uniqueness parameter to obtain the fitness value of each individual, randomly select multiple individuals from the initial individual population to compete, sort the individuals in the initial individual population according to the fitness value of each individual, select the target individual with a higher ranking as the parent individual based on the set ratio, and for the selected parent individual, uniformly exchange the corresponding vector element position with a first set probability to generate a new child individual, and generate vector elements of the child individual by linearly combining the parent individual vector elements, and add random noise to the vector elements of the child individual according to the second set probability to achieve mutation, wherein the second set probability is dynamically adjusted according to the fitness value of the child individual, and the child individual with a low fitness value has a higher second set probability.

[0181] Step S167, decoding the offspring individuals generated by the genetic operation and converting them into processable feature representations.

[0182] Step S168, use principal component analysis to decompose the decoded processable feature representation, extract the main component features and independent component features, and screen and filter the decomposed main component features and independent component features according to the pre-set traffic congestion semantic rules to remove some component features that are irrelevant or weakly related to the traffic congestion semantics. Then, the screened component features are reorganized according to the target strategy, and the key components in the screened component features are enhanced through nonlinear transformation to obtain the final traffic congestion semantic enhancement features.

[0183] Step S169: Load the traffic congestion semantic enhancement feature into the fully connected mapping unit, and generate the target traffic congestion knowledge point data for each traffic congestion knowledge point.

[0184] In this embodiment, it is assumed that the server has received target road traffic perception flow data for the past month from the traffic data management center. These target road traffic perception flow data cover the traffic conditions of multiple main roads in the city at different time periods, including detailed information such as traffic flow, vehicle speed, and vehicle queue length.

[0185] First, the server uses a Gaussian filter algorithm to remove noise from the target road traffic perception stream and generate a noise-cleaned perception stream. For example, in a busy commercial street perception stream, the original data may contain abnormal fluctuations and noise caused by equipment failure or environmental interference. Through Gaussian filtering, the server smooths out these noises, making the data more reflective of the actual traffic flow change trend.

[0186] Then, the contrast of the noise-cleaned perception stream is adjusted based on the preset contrast adjustment strategy to highlight the key elements in the traffic scene corresponding to the noise-cleaned perception stream and generate a preprocessed perception stream. For example, during the afternoon rush hour on a weekday, for the perception stream of a main road, the server enhances the contrast of the parts with heavy traffic and slow speed, making the traffic congestion more obvious and prominent.

[0187] Next, the server uses multiple convolutional layers in the convolutional neural network to capture different features in the preprocessed perception stream. Assume that the first layer of the convolutional layer captures the overall trend of traffic flow, the second layer captures the changing pattern of vehicle speed, and the third layer captures the distribution characteristics of the length of vehicle queues, etc. And use activation functions to perform nonlinear transformations on the captured different features, such as using the ReLU activation function to set negative values ​​to zero and retain positive values ​​to introduce nonlinearity and better fit complex traffic patterns.

[0188] After that, the server uses the maximum pooling operation to downsample the output of the last convolutional layer and uses the output of the last convolutional layer as the preliminary feature representation. For example, for a road perception flow in a specific area, the server extracts the most significant traffic features through the maximum pooling operation, reducing the amount of data while retaining the main information.

[0189] Next, a global average pooling operation is performed on the preliminary feature representation to obtain the average value of each feature channel. For example, for the three feature channels of traffic flow, vehicle speed, and vehicle queue length, the server calculates their respective average values.

[0190] Then, the server uses the attention mechanism to calculate the weight of each feature channel and uses the softmax function to normalize the weight of each feature channel to obtain the attention weight of each feature channel. Assuming that the average value of the traffic flow feature channel is large, it means that the channel has a greater impact on traffic congestion, and the server will give it a higher attention weight.

[0191] The server fuses the preliminary feature representation with the attention weight of each corresponding feature channel to obtain the weighted feature representation, and fuses the weighted feature representation to obtain the traffic-aware attention feature. For example, for a complex intersection, the server can highlight the traffic flow and vehicle queue length, two features that have a greater impact on traffic congestion, through the fusion operation.

[0192] The server loads the traffic-aware attention features into the semantic encoding unit to generate semantic encoding results. For example, the traffic-aware attention features are converted into a series of encoding vectors, each of which represents a specific traffic semantics.

[0193] Next, the server uses the semantic encoding result from the semantic encoding unit as the input of the traffic congestion semantic enhancement unit to generate an initial individual population. Assume that there are 100 individuals in the initial individual population, and each individual is a unique transformation representation of the input semantic encoding result.

[0194] The server calculates the feature correlation parameter between the transformed representation features generated by each individual and the known semantic features of traffic congestion, as well as the feature uniqueness parameter of each individual. For example, for an individual, if its transformed representation features are highly correlated with common traffic congestion patterns (such as main road congestion during peak hours), then its feature correlation parameter will be high. At the same time, if the transformed representation of this individual is significantly different from other individuals in the initial individual population, its feature uniqueness parameter will also be high.

[0195] The server performs weighted fusion of the feature correlation parameter and the feature uniqueness parameter to obtain the fitness value of each individual. For example, the weight of the feature correlation parameter is 0.7, and the weight of the feature uniqueness parameter is 0.3. The comprehensive fitness value of each individual is obtained through weighted calculation.

[0196] Then, the server randomly selects multiple individuals from the initial individual population to compete, and sorts the individuals in the initial individual population according to the fitness value of each individual. Suppose 50 individuals are selected to compete, and they are sorted from high to low according to the fitness value.

[0197] Based on the set ratio (e.g., the top 20%), the top ranked target individuals are selected as parent individuals. For the selected parent individuals, uniformly exchange the corresponding vector element positions with a first set probability (e.g., 0.5) to generate new child individuals, and generate the vector elements of the child individuals by linearly combining the parent individual vector elements. For example, for the vectors of two parent individuals, randomly exchange elements at certain positions, and generate the vector of the child individuals by linear combination.

[0198] The server adds random noise to the vector elements of the offspring individuals according to the second set probability to achieve mutation. Assuming that the lower the fitness value of the offspring individual, the higher the second set probability (for example, the offspring individual with a fitness value lower than the average value has a second set probability of 0.3; the offspring individual with a fitness value higher than the average value has a second set probability of 0.1).

[0199] Afterwards, the server decodes the offspring individuals generated by the genetic operation and converts them into processable feature representations, for example, converting the encoding vector of the offspring individuals into specific traffic feature values.

[0200] The server uses principal component analysis to decompose the decoded processable feature representation, extracts the main component features and independent component features, and screens and filters the decomposed main component features and independent component features according to the pre-set traffic congestion semantic rules to remove some component features that are irrelevant or weakly related to the traffic congestion semantics. For example, remove those minor features that have little impact on judging traffic congestion.

[0201] Finally, the server reorganizes the filtered component features according to the target strategy, and strengthens the key components in the filtered component features through nonlinear transformation to obtain the final traffic congestion semantic enhancement features.

[0202] The server loads the semantic enhancement features of traffic congestion into the fully connected mapping unit and generates target traffic congestion knowledge point data for each traffic congestion knowledge point. For example, for the traffic congestion knowledge point "between 8 and 9 in the morning on weekdays, the main road in the city center and connecting the main commercial district has a traffic volume of more than 1,500 vehicles per hour, an average speed of less than 15 kilometers per hour, and the length of the queue between two adjacent intersections exceeds 500 meters, and this state lasts for at least 30 minutes", the server generates the corresponding target traffic congestion knowledge point data to determine whether the current target road traffic perception flow meets the congestion situation described in this knowledge point.

[0203] In this way, the traffic perception flow of the target road can be deeply analyzed and processed, and its traffic congestion characteristics can be accurately judged whether they meet the preset traffic congestion knowledge points, providing strong support for traffic management and planning.

[0204] Figure 2 The hardware structure of the road traffic analysis system 100 based on edge computing for implementing the road traffic analysis method based on edge computing provided in an embodiment of the present application is shown as follows: Figure 2 As shown, the road traffic analysis system 100 based on edge computing may include a processor 110 , a machine-readable storage medium 120 , a bus 130 , and a communication unit 140 .

[0205] In one possible design, the road traffic analysis system 100 based on edge computing can be a single server or a server group. The server group can be centralized or distributed (for example, the road traffic analysis system 100 based on edge computing can be a distributed system). In some embodiments, the road traffic analysis system 100 based on edge computing can be local or remote. For example, the road traffic analysis system 100 based on edge computing can access information and / or data stored in a machine-readable storage medium 120 via a network. For another example, the road traffic analysis system 100 based on edge computing can be directly connected to the machine-readable storage medium 120 to access the stored information and / or data. In some embodiments, the road traffic analysis system 100 based on edge computing can be implemented on a road traffic analysis system based on edge computing. By way of example only, the road traffic analysis system based on edge computing can include a private cloud, a public cloud, a hybrid cloud, a community cloud, a distributed cloud, an internal cloud, a multi-layer cloud, etc., or any combination thereof.

[0206] The machine-readable storage medium 120 may store data and / or instructions. In some embodiments, the machine-readable storage medium 120 may store data obtained from an external terminal. In some embodiments, the machine-readable storage medium 120 may store data and / or instructions that the edge computing-based road traffic analysis system 100 uses to execute or use to complete the exemplary method described in this application.

[0207] During the specific implementation process, one or more processors 110 execute computer executable instructions stored in the machine-readable storage medium 120, so that the processor 110 can execute the road traffic analysis method based on edge computing in the above method embodiment. The processor 110, the machine-readable storage medium 120 and the communication unit 140 are connected through the bus 130, and the processor 110 can be used to control the sending and receiving actions of the communication unit 140.

[0208] The specific implementation process of the processor 110 can refer to the various method embodiments executed by the above-mentioned edge computing-based road traffic analysis system 100. The implementation principles and technical effects are similar, and this embodiment will not be repeated here.

[0209] In addition, an embodiment of the present application also provides a readable storage medium, in which computer executable instructions are preset. When the processor executes the computer executable instructions, the road traffic analysis method based on edge computing as described above is implemented.

[0210] It should be noted that in order to simplify the description disclosed in this application and thus help understand one or more embodiments of the invention, in the foregoing description of the embodiments of the present application, multiple features are sometimes combined into one embodiment, drawings, or descriptions thereof. Similarly, it should be noted that in order to simplify the description disclosed in this application and thus help understand one or more embodiments of the invention, in the foregoing description of the embodiments of the present application, multiple features are sometimes combined into one embodiment, drawings, or descriptions thereof.

Claims

1. A road traffic analysis method based on edge computing, characterized in that: The method comprises: Obtain road traffic data uploaded in real time in a traffic monitoring network; the road traffic data includes X traffic observation points, and the X traffic observation points all record data under the triggering of a first traffic flow update activity; the traffic observation points include road section traffic flow information corresponding to the traffic monitoring network; X is an integer greater than 1, and an edge computing device is deployed at each traffic observation point in the traffic monitoring network; Under the triggering of the first traffic flow update activity, collecting and recording the data of the X traffic flow observation points; and, Obtain edge computing data corresponding to the first traffic flow update activity generated by edge computing processing performed by an edge computing device deployed at each traffic flow observation point on the data of the traffic flow observation point, where the edge computing data corresponding to the first traffic flow update activity represents a real-time traffic status update description of the X traffic flow observation points; If the edge computing data corresponding to the first traffic flow update activity represents that the real-time traffic status update descriptions of the X traffic observation points are all valid status descriptions, then a data cloud storage configuration related to the first traffic flow update activity is performed in the traffic data management center corresponding to the traffic monitoring network; the data cloud storage configuration related to the first traffic flow update activity represents that the edge computing records of the traffic flow data collected by the X traffic observation points are stored in the traffic data management center; The method further comprises: Acquire a target road traffic perception flow described by a set type based on edge computing records of traffic flow data within a set time period in the traffic data management center in real-time traffic status update description; The target road traffic perception flow is loaded into the traffic congestion knowledge point description network, and target traffic congestion knowledge point data is obtained for each traffic congestion knowledge point, wherein the target traffic congestion knowledge point data indicates whether the traffic congestion characteristics of the target road traffic perception flow conform to the traffic congestion knowledge point; The traffic congestion knowledge point description network includes a semantic encoding unit, a traffic congestion semantic enhancement unit and a fully connected mapping unit; The step of loading the target road traffic perception flow into the traffic congestion knowledge point description network and obtaining target traffic congestion knowledge point data for each traffic congestion knowledge point includes: A Gaussian filtering algorithm is used to remove noise in the target road traffic perception stream to generate a noise-cleaned perception stream, and the contrast of the noise-cleaned perception stream is adjusted based on a preset contrast adjustment strategy to highlight key elements in the traffic scene corresponding to the noise-cleaned perception stream to generate a preprocessed perception stream; A multi-layer convolutional layer in a convolutional neural network is used to capture different features in the preprocessed perception stream, and an activation function is used to perform nonlinear transformation on the captured different features, and then a maximum pooling operation is used to downsample the output of the last convolutional layer, and the output of the last convolutional layer is used as a preliminary feature representation, and a global average pooling operation is performed on the preliminary feature representation to obtain the average value of each feature channel, and the weight of each feature channel is calculated using an attention mechanism according to the average value of each feature channel, and the weight of each feature channel is normalized using a softmax function to obtain the attention weight of each feature channel; fusing the preliminary feature representation with the attention weight of each corresponding feature channel to obtain a weighted feature representation, and fusing the weighted feature representation to obtain a traffic perception attention feature; Loading the traffic perception attention feature into the semantic coding unit to generate a semantic coding result, using the semantic coding result from the semantic coding unit as an input of the traffic congestion semantic enhancement unit to generate an initial individual population, wherein each individual in the initial individual population is a transformed representation of the input semantic coding result; Calculating a feature correlation parameter between the transformed representation feature generated by each individual and the known semantic feature of traffic congestion and a feature uniqueness parameter of each individual, wherein the feature uniqueness is used to characterize the degree of difference from the transformed representation features generated by other individuals in the initial individual population; The feature correlation parameter and the feature uniqueness parameter are weightedly fused to obtain the fitness value of each individual; A plurality of individuals are randomly selected from the initial individual population for competition, the individuals in the initial individual population are sorted according to the fitness value of each individual, a target individual with a higher sorting is selected as a parent individual based on a set ratio, and for the selected parent individual, a uniform exchange is performed at a corresponding vector element position with a first set probability to generate a new child individual, and a vector element of the child individual is generated by linearly combining the vector elements of the parent individual; Adding random noise to the vector elements of the offspring individuals according to a second set probability to achieve mutation, wherein the second set probability is dynamically adjusted according to the fitness value of the offspring individuals, and the offspring individuals with low fitness values ​​have a higher second set probability; Decode the offspring individuals generated by genetic operations and convert them into processable feature representations; The decoded processable feature representation is decomposed using principal component analysis to extract the main component features and independent component features. The decomposed main component features and independent component features are screened and filtered according to the pre-set traffic congestion semantic rules to remove some component features that are irrelevant or weakly related to the traffic congestion semantics. The screened component features are reorganized according to the target strategy, and the key components in the screened component features are enhanced through nonlinear transformation to obtain the final traffic congestion semantic enhancement features. The traffic congestion semantic enhancement feature is loaded into the fully connected mapping unit, and the target traffic congestion knowledge point data is generated for each traffic congestion knowledge point.

2. The road traffic analysis method based on edge computing according to claim 1 is characterized in that: The method further comprises: If the edge computing data corresponding to the first traffic flow update activity indicates that there is a traffic observation point whose real-time traffic status update description is an invalid status description among the X traffic observation points, then the data cloud storage related to the first traffic flow update activity is revoked in the traffic data management center corresponding to the traffic monitoring network; the data cloud storage revocation related to the first traffic flow update activity indicates that the edge computing records of the traffic flow data collected by the X traffic observation points are revoked in the traffic data management center; Among them, after the edge computing records of the traffic flow data collected by the X traffic observation points are configured to the traffic data management center by cloud storage, the traffic management service node can retrieve the traffic flow observation data related to the X traffic observation points from the traffic data management center; the edge computing records of the traffic flow data collected by the X traffic observation points also include: new records of adding the traffic flow observation data related to the X traffic observation points to the traffic data management center; revocation records of revoking the traffic flow observation data related to the X traffic observation points from the traffic data management center; adjustment records of adjusting the traffic observation data related to the X traffic observation points; and search records for the traffic observation data related to the X traffic observation points in the traffic data management center.

3. The road traffic analysis method based on edge computing according to claim 1 is characterized in that: The method further comprises: Constructing data access rights for the road traffic data; executing a permission protection instruction on the traffic observation data related to the X traffic observation points in the data collection process according to the data access rights, and the traffic observation data on which the permission protection instruction is executed is in a permission protection state; When the traffic observation data related to the X traffic observation points are in an authority protection state, the data of the X traffic observation points are collected and recorded under the triggering of the first traffic flow update activity.

4. The road flow analysis method based on edge computing according to claim 3 is characterized in that: The step of constructing data access rights for the road traffic data includes: Obtain a traffic data mapping bitmap of a traffic data management center corresponding to the traffic monitoring network; the traffic data mapping bitmap is used to record data index information of at least one road traffic data, wherein the at least one road traffic data includes the road traffic data uploaded in real time; Determining data index information of the road traffic data uploaded in real time from the traffic data mapping bitmap; The data access permission is configured for the road traffic data uploaded in real time according to the data index information of the road traffic data uploaded in real time.

5. The road traffic analysis method based on edge computing according to claim 3 is characterized in that: The step of constructing data access rights for the road traffic data includes: Obtain a traffic data mapping bitmap of a traffic data management center corresponding to the traffic monitoring network; the traffic data mapping bitmap is used to record data index information of at least one road traffic data, wherein the at least one road traffic data includes the road traffic data uploaded in real time; The data access permission is configured for the traffic data mapping bitmap; wherein the road traffic data corresponding to each data index information in the traffic data mapping bitmap is configured with the data access permission.

6. The road flow analysis method based on edge computing according to claim 3 is characterized in that: Under the condition that the data cloud storage configuration related to the first traffic flow update activity is performed in the traffic data management center corresponding to the traffic monitoring network, or under the condition that the data cloud storage related to the first traffic flow update activity is revoked in the traffic data management center corresponding to the traffic monitoring network, the method further includes: The data access right is released for the permission protection instruction of the traffic observation data related to the X traffic observation points; after the permission protection instruction is released, the traffic observation data related to the X traffic observation points is configured in a modifiable state.

7. The road traffic analysis method based on edge computing according to claim 1 is characterized in that: The road traffic data also includes Y traffic observation points, the recording time sequence of the Y traffic observation points is located after the X traffic observation points, and the Y traffic observation points all record data under the triggering of the second traffic flow update activity; Y is an integer greater than 1; the method further comprises: Under the triggering of the second traffic flow update activity, recording the traffic flow data of the Y traffic flow observation periods; Performing edge computing processing on the data collection process corresponding to the second traffic flow update activity to generate edge computing data corresponding to the second traffic flow update activity; the edge computing data corresponding to the second traffic flow update activity represents a real-time traffic status update description of the Y traffic observation points; If the edge computing data corresponding to the second traffic flow update activity represents that the real-time traffic status update descriptions of the Y traffic observation points are all valid status descriptions, then a data cloud storage configuration related to the second traffic flow update activity is performed in the traffic data management center corresponding to the traffic monitoring network; the data cloud storage configuration related to the second traffic flow update activity represents the edge computing records of the traffic flow data of the Y traffic observation periods stored in the traffic data management center; The trigger requirement for terminating the recording of the road traffic data is that the traffic flow data of the last traffic observation point in the road traffic data is completely recorded.

8. The road traffic analysis method based on edge computing according to claim 1 or 3, characterized in that: In the data collection process, each of the X traffic observation points is configured with a corresponding data record entry, and the data record entry covers the road section traffic flow information corresponding to the traffic monitoring network contained in the corresponding traffic observation point; there is a data record entry corresponding to the traffic observation point among the X traffic observation points, and the data processing module is associated with the data processing module, and the data processing module covers the priori road section traffic flow information associated with the corresponding traffic observation point, and the priori road section traffic flow information is the road section traffic flow information constructed by priori knowledge; after the data cloud storage configuration related to the first traffic flow update activity is performed in the traffic data management center corresponding to the traffic monitoring network, it also includes: Activates the data processing module associated with the data log entry.

9. The road traffic analysis method based on edge computing according to claim 8 is characterized in that: Z data record entries are associated with data processing modules, where Z is a positive integer and Z≤X; activating the data processing modules associated with the data record entries includes: Performing distributed parallel processing on the data processing modules associated with the Z data record entries; Among them, the data record entry corresponding to the flow observation point is associated with the data processing module both when the entry is generated and when the data processing is completed; Among them, among the data processing modules associated with the Z data record entries, the data processing modules that are processed in distributed parallel include: a data processing module associated with each data record entry among the Z data record entries except for the data record entries belonging to the target type observation points when the entry is generated and when the data processing is completed, and a data processing module associated with the data record entries among the Z data record entries belonging to the target type observation points at the trigger node of the entry construction; a data processing module associated with the data record entries among the Z data record entries belonging to the target type observation points at the termination node of the entry construction, which is activated in the next traffic flow update activity of the first traffic flow update activity.

10. A road traffic analysis system based on edge computing, characterized in that: The road traffic analysis system based on edge computing includes a processor and a memory, the memory is connected to the processor, the memory is used to store programs, instructions or codes, and the processor is used to execute the programs, instructions or codes in the memory to implement the road traffic analysis method based on edge computing as described in any one of claims 1 to 9 above.