Dynamic and static traffic cooperative treatment method and system

Abstract

The invention belongs to the technical field of traffic management, and particularly relates to a dynamic and static traffic cooperative governance method and system, and the method comprises the steps: collecting traffic dynamic and static data, and uploading the data to a traffic basic platform for data processing, network transmission, safety protection, privacy protection and data storage; the traffic dynamic and static data are input into the data platform for data arrangement, the artificial intelligence platform processes the data from different ports, the traffic data are deeply analyzed and predicted by using an intelligent algorithm, and then different traffic problems are optimized and adjusted by using the integrated algorithm platform; finally, the algorithm is applied to various traffic scenes, real-time data and analysis results of urban traffic are visually displayed through an intelligent large screen, and efficient management and optimization of an urban traffic system are achieved through the functions of multi-source data integration, intelligent algorithm support, large-scale data processing and simulation reasoning. And technical support and collaborative optimization are provided for constructing a more intelligent, efficient and safe intelligent traffic system.

Description

Technical Field

[0001] The present invention belongs to the technical field of traffic management, and in particular relates to a method and system for collaborative management of dynamic and static traffic. Background Art

[0002] With the acceleration of global urbanization and the continuous growth of urban population, traffic pressure is increasing. Especially in large and medium-sized cities, traffic congestion has become a major obstacle to the quality of life of residents and the development of urban economy. Traditional traffic management methods, such as relying solely on time control of traffic lights, fixed route planning and simple traffic flow monitoring, are no longer able to cope with the complex and changeable traffic conditions in modern cities. Smart transportation uses information and communication technology to efficiently allocate traffic resources, realize functions such as traffic flow optimization, signal control, and vehicle scheduling, and greatly improves the operating efficiency of urban transportation. However, the realization of smart transportation faces many challenges, especially in how to effectively integrate and process a large amount of dynamic and static traffic data. The complexity, diversity and real-time requirements of traffic data make data collection, transmission, fusion, processing and analysis extremely difficult.

[0003] In current traffic management, traditional methods usually rely on experience and rules, lacking flexibility and real-time response capabilities. Faced with dynamically changing traffic environments, traditional systems are often difficult to adjust and respond quickly, resulting in unsatisfactory traffic management results, which are specifically manifested in the dispersion and heterogeneity of traffic data, the dynamic nature of traffic information, the lack of global and intelligent optimization methods, and privacy and security issues. In response to the above challenges, research on smart transportation collaborative governance has gradually become a hot area. The core idea of ​​collaborative governance is to achieve comprehensive management and optimization of urban transportation systems through technical means, so that various traffic subsystems can work together to achieve greater overall effects. Summary of the invention

[0004] To achieve the above objectives, the present invention provides a method and system for collaborative management of dynamic and static traffic, which integrates dynamic and static traffic data, conducts real-time analysis and optimization of traffic, provides all-round support for urban traffic management, and improves the operating efficiency and intelligence level of urban traffic.

[0005] In order to achieve the above object, the present invention adopts the following technical solution: In a first aspect, the present invention provides a method for coordinated management of dynamic and static traffic, comprising the following steps: S1. Use the traffic terminal platform to collect traffic static data and traffic dynamic data; S2. Upload dynamic and static traffic data to the traffic infrastructure platform to realize data processing, network transmission, security protection, privacy protection and data storage of dynamic and static traffic data; S3. Input the dynamic and static traffic data stored in the traffic infrastructure platform into the data platform, and organize the data in the data platform, including data cleaning, data fusion and data processing; S4, the artificial intelligence platform trains deep learning models, intelligent reasoning and traffic scenario simulation for data from different ports, and uses intelligent algorithms to conduct in-depth analysis and prediction of traffic data; S5. Use the integrated algorithm platform to optimize and adjust different traffic problems; S6. Use the application platform to apply the algorithm to various traffic scenarios, and add smart large-screen displays to visualize the real-time data and analysis results of urban traffic.

[0006] As a further technical solution of the present invention, the traffic static data described in step S1 includes road network data, meteorological data, parking data, population distribution data and traffic facility data, wherein the road network data refers to the road structure of the city, including the number, width, road grade, intersection location, and shape of road intersections; meteorological data is provided in real time by a meteorological monitoring system and connected to a meteorological forecasting system to realize the prediction of future weather conditions; parking data includes the geographical location of the parking lot, the number of parking spaces, the real-time occupancy rate, and the real-time location, driving trajectory and speed information of the vehicle recorded by the vehicle GPS data; the population distribution data comes from the census data of the Statistics Bureau and the data analysis results of mobile signal operators; and the traffic facility data includes the distribution and status of traffic lights, signs, markings, guardrails, and isolation belts.

[0007] As a further technical solution of the present invention, the traffic dynamic data described in step S1 includes camera data, traffic flow data, traffic signal status data, navigation data, traffic event data and radar data; wherein the camera data is real-time dynamic video data obtained by a traffic checkpoint camera, the traffic flow data is the number of vehicles passing, speed, and density information obtained by the induction coils and traffic sensors on the road; the traffic signal status data is real-time monitoring of the switching status of traffic lights; the navigation data is collected by a vehicle navigation terminal to record the vehicle's route, speed and real-time position; the traffic event data includes traffic accidents, road construction, and traffic control conditions; the radar data is vehicle speed and distance information monitored by radar sensors.

[0008] As a further technical solution of the present invention, when step S2 performs network transmission, an edge computing and distributed network transmission architecture is adopted to ensure that dynamic and static data can be stably and quickly uploaded to the platform for processing in the case of massive collection and high-frequency updating. Edge computing can pre-process some data in advance by deploying computing power on edge nodes near data acquisition devices or sensors; a distributed data transmission method is adopted to divide the data into multiple data blocks and transmit them to the transportation infrastructure platform through different network paths.

[0009] As a further technical solution of the present invention, when uploading the static data described in step S2, a timed upload method is used for data transmission, and event-driven upload and time-driven upload methods are used for dynamic traffic data, and a data synchronization mechanism is added at the same time; when storing dynamic and static data, data encryption transmission is adopted, and the security and privacy of traffic data are ensured through access control and identity authentication and protection and desensitization of privacy data. At the same time, data verification and confirmation mechanism and redundant backup and storage are enabled to fully ensure efficient, stable and secure upload and storage of dynamic and static data.

[0010] As a further technical solution of the present invention, the data cleaning described in step S3 is to remove or correct erroneous, duplicate, incomplete or inconsistent data to ensure the accuracy and integrity of the data. The specific process is: first, a timestamp alignment method is used to ensure that the data collected by different devices and sensors are consistent in time, and then the data is deduplicated, duplicate records are identified and deleted, missing values ​​are filled with bilinear interpolation, and a logistic regression algorithm is used to correct abnormal data that may be measurement errors or transmission errors to ensure the accuracy and integrity of the data.

[0011] As a further technical solution of the present invention, the data fusion goal described in step S3 is to integrate these heterogeneous data together to form a consistent, structured data set. The specific process is: first, the data from different data sources are named by a field and the data type is standardized, and then the structured and unstructured data are fused using a weighted average algorithm. Finally, the time and space information of each data source is unified by combining timestamp correction and spatial coordinate standardization.

[0012] As a further technical solution of the present invention, the goal of the data processing described in step S3 is to perform in-depth processing and analysis on the cleaned and fused data. Specifically, a multi-layer perceptron is used to obtain preliminary features of the cleaned and fused data, and the original data and features are stored in a data warehouse.

[0013] As a further technical solution of the present invention, the artificial intelligence platform described in step S4 includes a training module, a reasoning module and a simulation module. The training module is responsible for training the deep learning model using historical data and real-time data so that the system can learn and predict traffic flow, vehicle paths, and congestion; the reasoning module is responsible for reasoning the real-time data based on the trained model to generate optimized traffic control decisions or future traffic trend predictions; the simulation module is responsible for conducting simulation experiments on various traffic scenarios. The artificial intelligence platform uses multi-source data to train a basic traffic model. The basic traffic model uniformly trains multi-source heterogeneous data and achieves excellent prediction results on multiple tasks. Specifically, variable tokenization and variable aggregation are used to perform preliminary feature extraction and modeling on variables in multi-source heterogeneous data, and on this basis, a unified basic traffic model is trained in a large amount of data using the Transformer model. The goal of the basic traffic model is to make predictions on multiple tasks using a unified model architecture, where the multi-source data is ,in represents a variable in the data, n represents the total number of variables, and the variable tokenization technology models each variable separately to obtain Encoded features ,in is a coding layer, where the coding layer is a linear layer, and then for each Perform variable aggregation operations to obtain the interactive information between the features of each variable and obtain the unified features of all variable information ,in represents the average pooling operation, is a feature concatenation operation, so contains the information of all variables, and then a mask matrix is ​​used to The features in the ,Will Input Transformer model to get the final features , and use the reconstruction loss function to optimize the model. The reconstruction loss function is: , where N is the number of all samples. At this time, the traffic basic model learns a unified feature representation from multi-source data. By adding a task head to the traffic basic model and fine-tuning it on a specific task, the traffic basic model is applied to different traffic scenarios.

[0014] As a further technical solution of the present invention, the integrated algorithm platform described in step S5 includes a traffic video detection algorithm, a flow prediction algorithm, a spatiotemporal analysis algorithm, a signal control algorithm, a location recommendation algorithm and a control algorithm. Each of the algorithms is obtained by fine-tuning the traffic basic model constructed in step S4, and collaborative optimization is achieved through cross-module collaboration and feedback on adjustment. Through collaborative optimization, the integrated algorithm platform can solve complex traffic management problems from multiple angles and dimensions, and improve the intelligence level and operation efficiency of the entire transportation system.

[0015] As a further technical solution of the present invention, the traffic prediction algorithm in step S5 obtains a unified traffic spatiotemporal representation Z through the trained traffic basic model. When the traffic basic model is applied to traffic flow prediction, a traffic prediction header is added to the end of the traffic basic model, and the data of the specific traffic prediction scene is used for full fine-tuning. Let the past traffic historical flow be X, the future traffic flow be Y, the traffic spatiotemporal representation obtained by the traffic basic model be Z, and the added traffic prediction header be expressed as: , is the traffic prediction value, Head is the task head added at the end, which is represented as a linear regression layer in the traffic prediction algorithm. Then, the MSE loss function is used to fine-tune the model. The MSE loss function is: .

[0016] As a further technical solution of the present invention, the traffic scenario described in step S6 includes traffic organization optimization, traffic flow prediction, traffic congestion diagnosis, signal control, integrated intelligence, vehicle OD analysis, vehicle analysis and vehicle control. With the addition of a smart large-screen display, the real-time data and analysis results of urban traffic can be visualized, providing an intuitive traffic monitoring interface for traffic managers. The application platform can provide timely and effective solutions through intelligent analysis and decision-making, and improve the operating efficiency of urban traffic. Through these optimization methods, city managers can better cope with the increasingly complex traffic environment and provide citizens with a more convenient and smooth travel experience.

[0017] In a second aspect, the present invention provides a dynamic and static traffic collaborative management system, comprising: The traffic terminal platform is used to collect dynamic and static traffic data from various sensors and devices, providing basic information for subsequent traffic management and optimization; The transportation infrastructure platform is used to realize data processing, network transmission, security protection, privacy protection and data storage of dynamic and static traffic data, providing powerful data processing and computing capabilities for the entire platform and ensuring data security and privacy; Data platform, used to clean, fuse and process dynamic and static traffic data stored in the traffic infrastructure platform, providing data support for intelligent analysis and reasoning; Artificial intelligence platform: The artificial intelligence platform trains deep learning models, intelligent reasoning and traffic scenario simulation for data from different ports, and uses intelligent algorithms to conduct in-depth analysis and prediction of traffic data; Integrated algorithm platform for optimization and adjustment of different traffic problems; The application platform is used to provide users with intuitive traffic application scenarios, visualize real-time data and analysis results of urban traffic, and realize collaborative management and optimization of the transportation system.

[0018] Compared with the prior art, the present invention has the following beneficial effects: By integrating multi-source traffic data, intelligent algorithms, simulation and reasoning capabilities, and modular architecture design, the present invention has achieved multiple improvements in data processing, real-time response, traffic optimization, and system expansion. First, dynamic and static traffic data are collected and integrated through the traffic end-side platform, and uniformly cleaned, processed, and integrated through the data platform to form a standardized data format. The comprehensive integration of multi-source data eliminates the disadvantages of data isolation and can provide a more complete information basis for traffic analysis and optimization. Moreover, the integration of static and dynamic data can provide support for long-term and short-term traffic management decisions, thereby improving the applicability of the platform at different time scales. Second, the modular design enhances the scalability of the system, can flexibly respond to future data growth and functional requirements, can quickly adapt to traffic management needs in different regions, realize customized deployment, and have strong compatibility. The dynamic and static traffic collaborative governance comprehensive platform realizes efficient management and optimization of the urban traffic system through multi-source data integration, intelligent algorithm support, large-scale data processing, and simulation and reasoning functions, and provides technical support and collaborative optimization for building a more intelligent, efficient, and safe smart transportation system. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] The accompanying drawings in the specification, which constitute a part of the present invention, are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute improper limitations on the present invention.

[0020] Figure 1 A flow chart of a method for collaborative management of dynamic and static traffic provided by the present invention.

[0021] Figure 2 A structural block diagram of a dynamic and static traffic collaborative management system provided by the present invention. DETAILED DESCRIPTION

[0022] The present invention will be further described below in conjunction with the accompanying drawings and embodiments.

[0023] It should be noted that the following detailed descriptions are exemplary and are intended to provide further explanation of the present invention. Unless otherwise specified, all technical and scientific terms used herein have the same meanings as those commonly understood by those skilled in the art to which the present invention belongs.

[0024] It should be noted that the terms used herein are only for describing specific embodiments, and are not intended to limit exemplary embodiments according to the present invention. As used herein, unless the context clearly indicates otherwise, the singular form is also intended to include the plural form. In addition, it should be understood that the terms "include" and "have" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0025] In the absence of conflict, the embodiments of the present invention and the features of the embodiments may be combined with each other.

[0026] Embodiment 1: like Figure 1 As shown, this embodiment provides a method for coordinated management of dynamic and static traffic, including the following steps: S1. Use the traffic end-side platform to collect static traffic data and dynamic traffic data; static traffic data includes road network data, meteorological data, parking data, population distribution data and traffic facility data, among which road network data refers to the road structure of the city, including the number, width, road grade, intersection location and shape of road intersections; meteorological data is provided in real time through the meteorological monitoring system and connected to the meteorological forecasting system to predict future weather conditions; parking data includes the geographical location of the parking lot, the number of parking spaces, the real-time occupancy rate and the real-time location, driving trajectory and speed information of the vehicle recorded by the vehicle GPS data; population distribution data comes from the census data of the Statistics Bureau and the data analysis results of mobile signal operators; traffic facility data includes The distribution and status of traffic lights, signs, markings, guardrails and isolation belts; traffic dynamic data includes camera data, traffic flow data, traffic signal status data, navigation data, traffic event data and radar data; among them, camera data is real-time dynamic video data obtained by traffic checkpoint cameras, traffic flow data is the number of vehicles passing, speed and density information obtained by induction coils and traffic sensors on the road; traffic signal status data is real-time monitoring of the switching status of traffic lights; navigation data is collected through vehicle navigation terminals, recording the vehicle's route, speed and real-time location; traffic event data includes traffic accidents, road construction and traffic control conditions; radar data is vehicle speed and distance information monitored by radar sensors.

[0027] S2. Upload dynamic and static traffic data to the traffic infrastructure platform to realize data processing, network transmission, security protection, privacy protection and data storage of dynamic and static traffic data; when performing network transmission, adopt edge computing and distributed network transmission architecture to ensure that dynamic and static data can be stably and quickly uploaded to the platform for processing in the case of massive collection and high-frequency update. Edge computing can pre-process some data in advance by deploying computing power on edge nodes near data collection devices or sensors; adopt distributed data transmission to divide data into multiple data blocks and transmit them to the traffic infrastructure platform through different network paths; when uploading static data, use scheduled upload to transmit data, and use event-driven upload and time-driven upload for dynamic traffic data, and add data synchronization mechanism; when storing dynamic and static data, use data encryption transmission, and ensure the security and privacy of traffic data through access control and identity authentication and protection and desensitization of privacy data. At the same time, enable data verification and confirmation mechanism and redundant backup and storage to fully ensure efficient, stable and secure upload and storage of dynamic and static data.

[0028] S3. Input the dynamic and static traffic data stored in the traffic infrastructure platform into the data platform, and organize the data in the data platform, including data cleaning, data fusion and data processing. Data cleaning is to remove or correct erroneous, duplicated, incomplete or inconsistent data to ensure the accuracy and integrity of the data. The specific process is: first, use the timestamp alignment method to ensure that the data collected by different devices and sensors are consistent in time, then deduplicate the data, identify and delete duplicate records, use bilinear interpolation to fill missing values, and use the logistic regression algorithm to correct abnormal data that may be measurement errors or transmission errors to ensure the accuracy and integrity of the data; the goal of data fusion is to integrate these heterogeneous data to form a consistent and structured data set. The specific process is: first, name the data from different data sources and normalize the data type, then use the weighted average algorithm to fuse the structured and unstructured data, and finally combine the timestamp correction and spatial coordinate standardization to unify the time and space information of each data source; the goal of data processing is to deeply process and analyze the cleaned and fused data. Specifically, for the cleaned and fused data, a multi-layer perceptron is used to obtain the preliminary features of the data, and the original data and features are stored in the data warehouse.

[0029] S4. The artificial intelligence platform trains deep learning models, intelligent reasoning and traffic scenario simulation for data from different ports, and uses intelligent algorithms to conduct in-depth analysis and prediction of traffic data; the artificial intelligence platform includes training modules, reasoning modules and simulation modules. The training module is responsible for training deep learning models using historical data and real-time data so that the system can learn and predict traffic flow, vehicle paths and congestion; the reasoning module is responsible for reasoning real-time data based on the trained model to generate optimized traffic control decisions or future traffic trend predictions; the simulation module is responsible for conducting simulation experiments on various traffic scenarios. The artificial intelligence platform uses multi-source data to train a basic traffic model. The basic traffic model will uniformly train multi-source heterogeneous data and achieve excellent prediction results on multiple tasks. Specifically, variable tokenization and variable aggregation are used to perform preliminary feature extraction and modeling on variables in multi-source heterogeneous data, and on this basis, a unified basic traffic model is trained in a large amount of data using the Transformer model. The goal of the basic traffic model is to use a unified model architecture to make predictions on multiple tasks, where multi-source data is ,in represents a variable in the data, n represents the total number of variables, and the variable tokenization technology models each variable separately to obtain Encoded features ,in is a coding layer. In this embodiment, the coding layer is a linear layer. Then, for each Perform variable aggregation operations to obtain the interactive information between the features of each variable and obtain the unified features of all variable information ,in represents the average pooling operation, is a feature concatenation operation, so contains the information of all variables, and then a mask matrix is ​​used to The features in the ,Will Input Transformer model to get the final features , and use the reconstruction loss function to optimize the model. The reconstruction loss function is: , where N is the number of all samples. At this time, the traffic basic model learns a unified feature representation from multi-source data. By adding a task head to the traffic basic model and fine-tuning it on a specific task, the traffic basic model is applied to different traffic scenarios.

[0030] S5. Optimize and adjust different traffic problems using an integrated algorithm platform; the integrated algorithm platform includes a traffic video detection algorithm, a flow prediction algorithm, a spatiotemporal analysis algorithm, a signal control algorithm, a location recommendation algorithm, and a control algorithm. Each algorithm is obtained by fine-tuning the traffic basic model constructed in step S4. Collaborative optimization is achieved through cross-module collaboration and feedback to the adjustment. Through collaborative optimization, the integrated algorithm platform can solve complex traffic management problems from multiple angles and dimensions, and improve the intelligence level and operation efficiency of the entire traffic system. This embodiment takes the flow prediction algorithm as an example, obtains a unified traffic spatiotemporal representation Z through a trained traffic basic model, and when the traffic basic model is applied to traffic flow prediction, a flow prediction header is added to the end of the traffic basic model, and the data of the specific traffic prediction scene is used for full fine-tuning. Let the past traffic historical flow be X, the future traffic flow be Y, and the traffic spatiotemporal representation obtained using the traffic basic model be Z. The added flow prediction header is represented as: , is the traffic prediction value, Head is the task head added at the end, which is represented as a linear regression layer in the traffic prediction algorithm and can have different representations in other tasks. Then, the MSE loss function is used to fine-tune the model. The MSE loss function is: ,By fine-tuning different application scenarios, the algorithms of different modules can be integrated into the ,algorithm process of the basic model, thus improving the intelligence level and ,operation efficiency of the entire transportation system.

[0031] S6. Use the application platform to apply the algorithm to various traffic scenarios, including traffic organization optimization, traffic flow prediction, traffic congestion diagnosis, signal control, integrated intelligence, vehicle OD analysis, vehicle analysis and vehicle control. Add smart large-screen display to visualize the real-time data and analysis results of urban traffic, and provide traffic managers with an intuitive traffic monitoring interface. The application platform can provide timely and effective solutions through intelligent analysis and decision-making, improve the operating efficiency of urban traffic, and through these optimization methods, city managers can better cope with the increasingly complex traffic environment and provide citizens with a more convenient and smooth travel experience.

[0032] Embodiment 2: like Figure 2 As shown, this embodiment provides a dynamic and static traffic collaborative management system, including: The traffic terminal platform is used to collect dynamic and static traffic data from various sensors and devices, providing basic information for subsequent traffic management and optimization; The transportation infrastructure platform is used to realize data processing, network transmission, security protection, privacy protection and data storage of dynamic and static traffic data, providing powerful data processing and computing capabilities for the entire platform and ensuring data security and privacy; Data platform, used to clean, fuse and process dynamic and static traffic data stored in the traffic infrastructure platform, providing data support for intelligent analysis and reasoning; Artificial intelligence platform: The artificial intelligence platform trains deep learning models, intelligent reasoning and traffic scenario simulation for data from different ports, and uses intelligent algorithms to conduct in-depth analysis and prediction of traffic data; Integrated algorithm platform for optimization and adjustment of different traffic problems; The application platform is used to provide users with intuitive traffic application scenarios, visualize real-time data and analysis results of urban traffic, and realize collaborative management and optimization of the transportation system.

[0033] In further embodiments, there is also provided: An electronic device includes a memory and a processor, and is equipped with computer instructions running on the processor. When these instructions are executed by the processor, the method steps described in the embodiment can be accurately implemented. For the sake of simplicity, the detailed method content will not be repeated here.

[0034] It should be clear that the processor in this embodiment is not limited to the central processing unit CPU, it can also be a general-purpose processor core, a digital signal processor DSP, an application-specific integrated circuit ASIC, a field programmable gate array FPGA or other programmable logic device PLD, a discrete gate circuit or transistor logic component, a discrete hardware module, etc. In addition, the general-purpose processor can be a microprocessor MPU, or any conventional computing core.

[0035] The memory may include a read-only memory and a random access memory, and provide instructions and data to the processor. A portion of the memory may also include a non-volatile random access memory. For example, the memory may also store information about the device type.

[0036] A computer-readable storage medium is used to store computer instructions. When the computer instructions are executed by a processor, the method described in the embodiment is completed.

[0037] The method in the embodiment can be directly embodied as being executed by a hardware processor, or being executed by a combination of hardware and software modules in the processor. The software module can be located in a mature storage medium in the art such as a random access memory, a flash memory, a read-only memory, a programmable read-only memory, or an electrically erasable programmable memory, a register, etc. The storage medium is located in a memory, and the processor reads the information in the memory and completes the steps of the above method in combination with its hardware. To avoid repetition, it will not be described in detail here.

[0038] Those skilled in the art will appreciate that the units, i.e., algorithm steps, of the various examples described in conjunction with this embodiment can be implemented in electronic hardware or in a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0039] The above are only preferred embodiments of the present invention and are not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

[0040] Although the above describes the specific implementation mode of the present invention in conjunction with the accompanying drawings, it is not intended to limit the scope of protection of the present invention. Those skilled in the art should understand that various modifications or variations that can be made by those skilled in the art on the basis of the technical solution of the present invention without creative work are still within the scope of protection of the present invention.

Claims

1. A method for coordinated management of dynamic and static traffic, characterized in that: The following steps are involved: S1. Use the traffic terminal platform to collect traffic static data and traffic dynamic data; S2. Upload dynamic and static traffic data to the traffic infrastructure platform to realize data processing, network transmission, security protection, privacy protection and data storage of dynamic and static traffic data; S3. Input the dynamic and static traffic data stored in the traffic infrastructure platform into the data platform, and organize the data in the data platform, including data cleaning, data fusion and data processing; S4, the artificial intelligence platform trains deep learning models, intelligent reasoning and traffic scenario simulation for data from different ports, and uses intelligent algorithms to conduct in-depth analysis and prediction of traffic data; S5. Use the integrated algorithm platform to optimize and adjust different traffic problems; S6. Use the application platform to apply the algorithm to various traffic scenarios, and add smart large-screen displays to visualize the real-time data and analysis results of urban traffic.

2. The method for coordinated management of dynamic and static traffic according to claim 1 is characterized in that: The traffic static data in step S1 includes road network data, meteorological data, parking data, population distribution data and traffic facility data, wherein the road network data refers to the road structure of the city, including the number, width, road grade, intersection location and shape of road intersections; the meteorological data is provided in real time by the meteorological monitoring system and connected with the meteorological forecasting system to realize the prediction of future weather conditions; Parking data includes the geographic location of the parking lot, the number of parking spaces, the real-time occupancy rate, and the real-time location, driving trajectory and speed information of the vehicle recorded by the vehicle GPS data; population distribution data comes from the census data of the Statistics Bureau and the data analysis results of mobile signal operators; traffic facility data includes the distribution and status of traffic lights, signs, markings, guardrails, and isolation belts.

3. The method for coordinated management of dynamic and static traffic according to claim 2 is characterized in that: The traffic dynamic data in step S1 includes camera data, traffic flow data, traffic signal status data, navigation data, traffic event data and radar data; The camera data is the real-time dynamic video data obtained by the traffic checkpoint camera; the traffic flow data is the number of vehicles passing, speed, and density information obtained by the induction coils and traffic sensors on the road; the traffic signal status data is the real-time monitoring of the switching status of traffic lights; the navigation data is collected by the vehicle navigation terminal, recording the vehicle's driving route, driving speed, and real-time location; the traffic event data includes traffic accidents, road construction, and traffic control conditions; Radar data is vehicle speed and distance information monitored by radar sensors.

4. The method for coordinated management of dynamic and static traffic according to claim 3 is characterized in that: When performing network transmission in step S2, the architecture of edge computing and distributed network transmission is adopted to ensure that dynamic and static data can be stably and quickly uploaded to the platform for processing in the case of massive collection and high-frequency updating. Edge computing can pre-process some data in advance by deploying computing power on edge nodes near data collection devices or sensors; distributed data transmission is adopted to divide the data into multiple data blocks and transmit them to the transportation infrastructure platform through different network paths.

5. The method for coordinated management of dynamic and static traffic according to claim 4 is characterized in that: When uploading the static data described in step S2, the data is transmitted by timed uploading, and the dynamic traffic data is uploaded by event-driven uploading and time-driven uploading, and a data synchronization mechanism is added at the same time; when storing dynamic and static data, data encryption transmission is adopted, and the security and privacy of traffic data are ensured through access control and identity authentication and protection and desensitization of privacy data, and data verification and confirmation mechanism and redundant backup and storage are enabled at the same time.

6. The method for coordinated management of dynamic and static traffic according to claim 5 is characterized in that: The data cleaning in step S3 is to remove or correct erroneous, duplicate, incomplete or inconsistent data. The specific process is: first, use the timestamp alignment method to ensure that the data collected by different devices and sensors are consistent in time, then deduplicate the data, identify and delete duplicate records, use bilinear interpolation to fill missing values, and use the logistic regression algorithm to correct abnormal data that may be measurement errors or transmission errors; the goal of data fusion is to integrate these heterogeneous data together to form a consistent and structured data set. The specific process is: first, the data from different data sources are named by a field and the data type is normalized, and then the structured and unstructured data are fused using a weighted average algorithm. Finally, the time and space information of each data source is unified by combining timestamp correction and spatial coordinate standardization; the goal of data processing is to deeply process and analyze the cleaned and fused data. Specifically, for the cleaned and fused data, a multi-layer perceptron is used to obtain the preliminary features of the data, and the original data and features are stored in a data warehouse.

7. The method for coordinated management of dynamic and static traffic according to claim 6 is characterized in that: The artificial intelligence platform described in step S4 includes a training module, a reasoning module and a simulation module. The training module is responsible for training the deep learning model using historical data and real-time data so that the system can learn and predict traffic flow, vehicle paths, and congestion; the reasoning module is responsible for reasoning the real-time data based on the trained model to generate optimized traffic control decisions or future traffic trend predictions; the simulation module is responsible for conducting simulation experiments on various traffic scenarios. The artificial intelligence platform uses multi-source data to train a basic traffic model. The basic traffic model will uniformly train multi-source heterogeneous data and achieve excellent prediction results on multiple tasks. Specifically, variable segmentation and variable aggregation are used to perform preliminary feature extraction and modeling on the variables in the multi-source heterogeneous data, and on this basis, the Transformer model is used to train a unified basic traffic model in a large amount of data. The goal of the basic traffic model is to use a unified model architecture to make predictions on multiple tasks, where the multi-source data is ,in represents a variable in the data, n represents the total number of variables, and the variable segmentation technology models each variable separately to obtain Encoded features ,in is a coding layer, the coding layer is a linear layer, and then for each Perform variable aggregation operations to obtain the interactive information between the features of each variable and obtain the unified features of all variable information ,in represents the average pooling operation, is a feature concatenation operation, so Contains information about all variables. At this time, a mask matrix is ​​used to The features are randomly masked to obtain ,Will Input Transformer model to get the final features , and use the reconstruction loss function to optimize the model. The reconstruction loss function is: , where N is the number of all samples. At this time, the traffic basic model learns a unified feature representation from multi-source data. By adding a task head to the traffic basic model and fine-tuning it on a specific task, the traffic basic model is applied to different traffic scenarios.

8. The method for coordinated management of dynamic and static traffic according to claim 7 is characterized in that: The integrated algorithm platform in step S5 includes a traffic video detection algorithm, a flow prediction algorithm, a spatiotemporal analysis algorithm, a signal control algorithm, a location recommendation algorithm and a control algorithm. Each of the algorithms is obtained by fine-tuning the traffic basic model constructed in step S4. The flow prediction algorithm obtains a unified traffic spatiotemporal representation Z through the trained traffic basic model. When the traffic basic model is applied to traffic flow prediction, a flow prediction header is added to the end of the traffic basic model, and the data of the specific scene of traffic prediction is used for full fine-tuning. Assume that the past traffic historical flow is X, the future traffic flow is Y, the traffic spatiotemporal representation obtained by using the traffic basic model is Z, and the added flow prediction header is expressed as: , is the traffic prediction value, Head is the task head added at the end, which is represented as a linear regression layer in the traffic prediction algorithm. Then, the MSE loss function is used to fine-tune the model. The MSE loss function is: .

9. According to the method for coordinated management of dynamic and static traffic in claim 8, the traffic scenario in step S6 includes traffic organization optimization, traffic flow prediction, traffic congestion diagnosis, signal control, integrated intelligence, vehicle OD analysis, vehicle analysis and vehicle control.

10. A dynamic and static traffic collaborative management system, characterized in that: The method according to any one of claims 1 to 9 comprises: The traffic terminal platform is used to collect dynamic and static traffic data from various sensors and devices, providing basic information for subsequent traffic management and optimization; The transportation infrastructure platform is used to realize data processing, network transmission, security protection, privacy protection and data storage of dynamic and static traffic data, providing powerful data processing and computing capabilities for the entire platform and ensuring data security and privacy; Data platform, used to clean, fuse and process dynamic and static traffic data stored in the traffic infrastructure platform, providing data support for intelligent analysis and reasoning; Artificial intelligence platform: The artificial intelligence platform trains deep learning models, intelligent reasoning and traffic scenario simulation for data from different ports, and uses intelligent algorithms to conduct in-depth analysis and prediction of traffic data; Integrated algorithm platform for optimization and adjustment of different traffic problems; The application platform is used to provide users with intuitive traffic application scenarios, visualize real-time data and analysis results of urban traffic, and realize collaborative management and optimization of the transportation system.

Images