Intelligent traffic flow dynamic regulation and control method and system based on multi-source data fusion

Abstract

The invention discloses an intelligent traffic flow dynamic regulation and control method and system based on multi-source data fusion, and relates to the technical field of data processing. The method comprises the following steps: executing multi-source data acquisition through a data acquisition interface to obtain a multi-source traffic data set; performing spatial fusion on the multi-source traffic data set to obtain a flow marking map; performing current traffic state analysis based on the flow marking map, identifying local and global congestion, and constructing a traffic situation map; and performing traffic flow state prediction in a preset future time zone based on the traffic situation map, performing regulation and control decision based on a prediction result, and generating a traffic flow regulation and control strategy. The technical problem that traffic flow prediction and regulation are not accurate enough in the prior art is solved, and the technical effects that intelligent traffic flow dynamic regulation is achieved through multi-source data fusion and space-time analysis, and the traffic management precision and the response speed are improved are achieved.

Description

Technical Field

[0001] The present invention relates to the field of data processing technology, and in particular to an intelligent traffic flow dynamic control method and system based on multi-source data fusion. Background Art

[0002] With the acceleration of urbanization and the increase in car ownership, urban traffic congestion is becoming increasingly serious. Traditional traffic flow management methods rely primarily on fixed traffic lights or simple flow analysis, which are unable to cope with complex and dynamically changing traffic conditions. Furthermore, existing traffic flow control systems mostly use a single data source for flow monitoring and prediction, failing to effectively integrate data from multiple sources such as traffic sensors, surveillance videos, and historical road data. This leads to limitations in control strategies and inaccuracies. Summary of the Invention

[0003] The present application provides a method and system for intelligent dynamic traffic flow control based on multi-source data fusion, which solves the technical problem of inaccurate traffic flow prediction and control in the prior art.

[0004] The first aspect of the present application provides a method for intelligent traffic flow dynamic control based on multi-source data fusion, the method comprising:

[0005] Multi-source data collection is performed through the data acquisition interface to obtain a multi-source traffic data set; the multi-source traffic data set is spatially fused to obtain a traffic mark map; the current traffic status is analyzed based on the traffic mark map, local and global congestion are identified, and a traffic situation map is constructed; based on the traffic situation map, traffic flow status prediction is performed in a preset future time zone, and control decisions are made based on the prediction results to generate a traffic flow control strategy.

[0006] The second aspect of the present application provides an intelligent traffic flow dynamic control system based on multi-source data fusion, the system comprising:

[0007] Data acquisition module: performs multi-source data acquisition through the data acquisition interface to obtain a multi-source traffic data set; spatial fusion module: performs spatial fusion on the multi-source traffic data set to obtain a traffic mark map; analysis module: performs current traffic status analysis based on the traffic mark map, identifies local and global congestion, and constructs a traffic situation map; control module: performs traffic flow status prediction in a preset future time zone based on the traffic situation map, makes control decisions based on the prediction results, and generates a traffic flow control strategy.

[0008] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0009] Multi-source data collection is performed through the data acquisition interface to obtain a multi-source traffic dataset. Next, the multi-source traffic dataset is spatially fused to obtain a traffic flow marker map. The current traffic status is then analyzed based on the traffic flow marker map, identifying local and global congestion and constructing a traffic situation map. Finally, based on the traffic situation map, traffic flow status predictions are performed within a preset future time zone. Traffic control decisions are made based on the prediction results, and a traffic flow control strategy is generated. This solves the technical problem of inaccurate traffic flow prediction and control in existing technologies, achieving the technical effect of intelligent dynamic traffic flow control through multi-source data fusion and spatiotemporal analysis, improving traffic management accuracy and response speed. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0011] Figure 1 A flow chart of a method for intelligent traffic flow dynamic control based on multi-source data fusion provided in an embodiment of the present application;

[0012] Figure 2 Schematic diagram of the structure of the intelligent traffic flow dynamic control system based on multi-source data fusion provided in the embodiment of the present application.

[0013] Description of the accompanying drawings: data acquisition module 11, spatial fusion module 12, analysis module 13, control module 14. DETAILED DESCRIPTION

[0014] This application solves the technical problem of inaccurate traffic flow prediction and control in the prior art by providing an intelligent traffic flow dynamic control method and system based on multi-source data fusion.

[0015] The following will be combined with the accompanying drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only some of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0016] It should be noted that the terms "including" and "having" are intended to cover non-exclusive inclusions. For example, a process, method, system, product or server that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or modules that are not clearly listed or are inherent to these processes, methods, products or devices.

[0017] Example 1, as Figure 1 As shown, the present application provides a method for intelligent traffic flow dynamic control based on multi-source data fusion, wherein the method includes:

[0018] Multi-source data collection is performed through the data collection interface to obtain a multi-source traffic data set.

[0019] In the embodiment of the present application, the data acquisition interface, as a key part of the system data input, undertakes the important task of uniformly accessing and collecting data from multiple types of data sources to form a multi-source traffic data set; the data acquisition interface is a standardized platform with the ability to be compatible with and access a variety of different data sources, covering various types of road perception devices (such as cameras, geomagnetic sensors, radars, etc.), vehicle-mounted systems (such as GPS navigation devices, vehicle-mounted diagnostic systems, etc.) and mobile terminals (such as navigation application data of users of smartphones, tablets, etc.), and has data format conversion and protocol parsing functions, which can convert the original data provided by different data sources into a unified format for subsequent processing and analysis; the data acquisition interface is connected via wired or wireless It accesses various sensing devices deployed along the road in a real-time manner to obtain basic data such as traffic flow, vehicle speed, and vehicle type, and interacts with the GPS navigation equipment and on-board diagnostic system on the vehicle to collect information such as the vehicle's real-time location, driving status, and driving behavior. It also obtains data such as travel intentions, route selection, and real-time location in the user's navigation application through an API interface or data push. After being processed by the data acquisition interface, data from different data sources are integrated into a unified multi-source traffic dataset, which contains rich traffic information such as road flow, speed distribution, congestion conditions, vehicle trajectories, and user travel needs, providing a solid data foundation for subsequent spatial fusion, traffic status analysis, and flow prediction.

[0020] Furthermore, the data acquisition interface is used to uniformly access multiple types of data sources, and the multiple types of data sources at least include various types of road perception devices, vehicle-mounted systems and mobile terminals.

[0021] The data acquisition interface centrally accesses and collects data from multiple disparate data sources, including various road sensing devices, in-vehicle systems, and mobile terminals. Specifically, road sensing devices such as pavement sensors, video surveillance cameras, radar, and lidar provide real-time data on road conditions, traffic flow, and vehicle speed. In-vehicle systems, including built-in GPS positioning, speed sensors, and accelerometers, provide data on the driving status and location of individual vehicles. Mobile terminals, on the other hand, capture user location, navigation, and traffic information via mobile devices such as smartphones. Particularly in urban areas, this information can provide real-time road condition and traffic feedback. By integrating these multiple data sources, the data acquisition interface comprehensively collects traffic data from various dimensions, providing accurate and comprehensive data support for subsequent traffic flow analysis and dynamic regulation.

[0022] The multi-source traffic dataset is spatially fused to obtain a traffic mark map.

[0023] Comprehensive traffic analysis is facilitated by integrating traffic information from diverse data sources into a unified spatial framework. This process first requires aligning various traffic data types to the same time scale, ensuring that data from different sources have identical time stamps. Next, a road network model is constructed, and the aligned data is spatially processed by mapping them onto specific road links and intersections. Based on this, the mapped data is weighted and fused. Each data point is assigned a weight based on the reliability and accuracy of its source, which can be initialized and adjusted using a preset confidence level. Finally, through weighted fusion of multiple data sources, traffic flow characteristics such as volume, average speed, and congestion index are extracted to construct a traffic marker map. This map displays information such as traffic volume and speed at different road sections and intersections, providing essential spatial information for subsequent traffic situation analysis and the development of traffic control strategies.

[0024] Furthermore, spatial fusion is performed on the multi-source traffic dataset to obtain a traffic mark map, including:

[0025] The multi-source traffic datasets are aligned to the same time scale to generate an aligned dataset; a road network model is constructed, the aligned datasets are mapped to road links and intersections, and data confidence-weighted fusion is performed to obtain a fusion result; traffic flow features are extracted and marked using the fusion result, including vehicle volume, average speed, and congestion index, to generate the traffic marked map.

[0026] Specifically, to address the temporal asynchrony caused by differences in collection frequency and timing between different data sources, techniques such as interpolation, extrapolation, or time window aggregation are used to align the time scales of multi-source traffic datasets. This generates an aligned dataset with a unified time base, laying the foundation for temporal consistency for subsequent spatial fusion. Subsequently, a road network model is constructed that reflects the actual road network topology. This model is presented as a graph structure, with nodes representing intersections, junctions, and pre-set landmarks, and directed edges corresponding to road segments. Accurate data mapping is ensured by accurately characterizing road geometry, direction, and length. After the road network model is constructed, the aligned dataset is mapped to the road links and intersections in the model based on spatial location information (such as GPS coordinates), ensuring a precise correspondence between the data and the actual roads. Furthermore, given the differences in data quality, coverage, and real-time performance among different data sources, reliability weights for the various data sources must be initialized to provide a weighting basis for subsequent weighted fusion. Based on initialization weights, a weighted fusion is performed on multi-source data mapped to road links and intersections, integrating information from different data sources to improve data accuracy and reliability. Anomalies such as GPS drift are detected and processed to ensure that the fusion results are not affected by abnormal data. Finally, key traffic flow characteristics such as vehicle volume, average speed, and congestion index are extracted from the fused data. These characteristics are then labeled on the road network model to generate a flow-labeled map that intuitively displays traffic flow, speed, and congestion conditions for each road segment.

[0027] Furthermore, the road network model is a graph structure model of a preset road area, wherein nodes are intersections, cross points and preset landmarks; and edges are directed edges, representing roads.

[0028] The road network model is a graph structured around a predefined road area. Within this model, nodes represent key locations within the traffic network, such as intersections, junctions, and predefined landmarks. Intersections and junctions, as key nodes in the road network, are crucial points where vehicles change direction or converge. Predefined landmarks are selected based on actual traffic management needs and road characteristics.

[0029] Edges in a road network model are directed edges, representing road connections, specifically the connections between intersections and junctions. These directed edges indicate vehicle travel directions, ensuring the correctness and logic of traffic flow. In a road network model, information about the traffic flow and operating conditions of each road is conveyed through these directed edges. For example, the direction of an edge indicates the flow of vehicles from one node to another, and the weight of an edge can represent traffic flow characteristics such as road volume, speed, and traffic density.

[0030] Furthermore, the aligned dataset is mapped to road links and intersections, and data confidence weighted fusion is performed to obtain a fusion result, including:

[0031] Initializing the reliability weights of multiple types of data sources connected to the data acquisition interface to generate initialization weights; performing GPS drift processing on the aligned data set, and then performing weighted fusion on the multi-source data mapped to the road links and intersections using the initialization weights to generate the fusion result.

[0032] Reliability weight initialization is carried out for various data sources accessed through the data collection interface. Optionally, a multi-dimensional evaluation system is constructed, covering core elements such as data quality, coverage, real-time performance and historical performance. In terms of data quality, it is necessary to consider the accuracy, completeness and error-free rate of the data, and quantitatively evaluate it through means such as data verification and consistency checks; coverage focuses on the breadth and continuity of data in spatial and temporal dimensions, and is measured using methods such as geographic information analysis and time series analysis; real-time evaluation focuses on data update frequency and transmission delay, and is achieved through technical means such as timestamp comparison and network performance monitoring; historical performance evaluation comprehensively considers the historical stability of the data source, the frequency of abnormal records, etc., and is quantified through tools such as statistical analysis and machine learning prediction models. Based on the above multi-dimensional evaluation results, weight allocation algorithms such as the hierarchical analysis method and the entropy weight method are used to scientifically assign initialization weights to each data source to objectively reflect its reliability in the overall data system and provide an accurate quantitative basis for subsequent weighted fusion.

[0033] GPS drift processing can be implemented on time-aligned datasets, using advanced anomaly detection and filtering techniques for dynamic correction. For anomaly detection, statistical outlier detection algorithms (such as the Z-score and IQR methods) or machine learning-based classification models (such as isolation forests and support vector machines) can be used to monitor GPS data in real time, identifying and marking drifting data points. For filtering, dynamic estimation methods such as Kalman filtering and particle filtering are used, combined with prior knowledge such as vehicle kinematic models and map matching techniques, to accurately correct drift data and ensure that location information truly reflects actual traffic conditions.

[0034] After GPS drift processing is complete, the system performs a weighted fusion of multi-source data mapped to road links and intersections based on initial weights. This process uses weighted averaging, weighted voting, or a Bayesian-based fusion algorithm based on the initial weights of each data source to comprehensively calculate the data from different sources. By assigning greater weight to more reliable sources, the impact of low-quality data on the overall results is effectively suppressed, resulting in a more accurate and reliable fusion result.

[0035] Based on the traffic mark map, the current traffic status is analyzed, local and global congestion are identified, and a traffic situation map is constructed.

[0036] By interpreting the traffic flow characteristics of each road segment and intersection marked on the flow-marked map, local and global traffic congestion conditions can be identified. Local congestion generally refers to traffic flow exceeding normal limits at a single road segment or intersection, which may be caused by traffic bottlenecks such as traffic accidents, construction, or other reasons. Global congestion, on the other hand, involves a wider regional traffic flow problem, which may be caused by traffic congestion in multiple adjacent road segments or areas that affect each other.

[0037] Based on the identified local and global congestion information, a traffic situation map is constructed. The traffic situation map is a visual representation of the current traffic status formed by comprehensively analyzing the traffic conditions of each road section and intersection, combining real-time traffic data, historical data, weather conditions and other factors. Each node in the map represents a traffic control unit (such as a road section or intersection), and the edges represent the flow transmission relationship between each traffic unit. In the traffic situation map, the color, size or shape of the node can reflect the traffic conditions in the area. For example, red indicates severe congestion, orange indicates moderate congestion, and green indicates unobstructed congestion.

[0038] Based on the traffic situation map, traffic flow state prediction in a preset future time zone is performed, and control decisions are made based on the prediction results to generate a traffic flow control strategy.

[0039] In an embodiment of the present application, based on the constructed traffic situation map, the system further performs a traffic flow state prediction within a preset future time zone. Specifically, by analyzing the current traffic conditions of each node (such as a road section or intersection) in the traffic situation map, and combining historical data, weather forecasts, special events and other factors, a machine learning model (such as a random forest, support vector machine) is used to predict the traffic flow state within a certain time zone in the future. The real-time data in the current traffic situation map and relevant external factors (such as weather forecasts, large-scale event arrangements, etc.) are used as input and substituted into the trained prediction model to predict the traffic flow state of each road section within a preset future time zone (such as the next 1 hour, 3 hours, 6 hours, etc.). By predicting the traffic flow in the future time zone, the system can accurately predict potential traffic congestion or smooth traffic. Based on these prediction results, the system will take corresponding regulatory decisions according to different traffic flow situations. For example, when severe congestion is predicted in certain sections of the road, the system can decide to take measures in advance, such as adjusting the timing of traffic lights, setting up temporary traffic control, or directing vehicles to other alternative routes. Ultimately, based on these traffic control decisions, the system generates traffic flow control strategies and distributes them in real time to the traffic management system, which then executes the corresponding traffic flow control operations. These strategies include optimizing signal timing (adjusting the green light duration and cycle length for each phase), disseminating information (providing optimal driving routes to drivers through variable information boards and mobile apps), traffic control measures (such as setting up tidal lanes and implementing one-way traffic), and public transportation priority strategies (such as increasing bus frequencies and establishing dedicated bus lanes).

[0040] Furthermore, traffic flow state prediction within a preset future time zone is performed based on the traffic situation map, and traffic flow control decisions are made based on the prediction results to generate a traffic flow control strategy, including:

[0041] The target regulated road section is located according to the prediction result, where the target regulated road section is a road section where the congestion level is greater than a preset congestion threshold and the congestion duration is greater than a preset time threshold; the adjustable modes of the target regulated road section are collected, including one or more of a signal light timing optimization mode, a route induction mode, and a lane control mode; based on the adjustable modes, with the goal of reducing the congestion level, a flow control analysis of the target regulated road section is performed to generate the traffic flow control strategy.

[0042] After completing the traffic flow prediction, the system screens and analyzes the prediction results based on preset congestion and time thresholds. Specifically, a congestion threshold is set, such as when the predicted congestion index reaches or exceeds 0.8, the road section is considered congested. A time threshold is also set, such as when a section is continuously congested for 15 minutes or more, the section is identified as a target for regulation. This screening process pinpoints sections of road that are likely to experience severe and prolonged congestion in the future time zone. These sections will become the focus of subsequent regulation strategies.

[0043] For the located target control section, the system further collects its adjustable modes. Adjustable modes include but are not limited to signal light timing optimization mode, route induction mode and lane control mode. For the signal light timing optimization mode, the system communicates with the traffic signal control equipment around the target control section to obtain the current signal light timing plan, including parameters such as the green light time, red light time, and cycle length of each phase. At the same time, combined with the traffic flow prediction results and the road geometry characteristics of the target control section (such as the number of lanes, intersection layout, etc.), the principles of traffic engineering and optimization algorithms (such as genetic algorithms, particle swarm algorithms, etc.) are used to optimize and adjust the signal light timing plan and generate a new timing plan to achieve smooth traffic flow at the intersection and reduce vehicle waiting time.

[0044] In terms of route guidance, the system integrates with traffic information dissemination platforms such as navigation applications and variable information boards to access currently available channels and methods for disseminating guidance information. Based on traffic flow forecasts, the system analyzes traffic conditions around the target regulated road section and plans the optimal route to avoid congestion. This route information is pushed to the driver in real time via the navigation application and displayed on the variable information board, guiding the driver to choose the right route, thereby dispersing traffic flow and alleviating traffic pressure on the target regulated road section.

[0045] For lane control modes, the system evaluates the feasibility and effectiveness of implementing control measures such as tidal lanes, variable lanes, and bus lanes based on the traffic flow and congestion conditions of the target control section. For example, if it is predicted that traffic flow in one direction is high and in the other direction is low within a specific time period on the target control section, the system can recommend the installation of tidal lanes to improve road resource utilization by adjusting the lane direction. Alternatively, based on traffic flow prediction results, the system can dynamically adjust the use of variable lanes, such as setting some lanes as left-turn lanes or straight-ahead lanes to optimize traffic flow organization. At the same time, bus lanes can be set up when necessary to ensure priority for public transportation, improve the efficiency of public transportation, attract more people to choose public transportation, and reduce the flow of private cars.

[0046] After collecting the available adjustable modes for the target regulated road section, the system combines and optimizes these modes with the core goal of reducing congestion. By establishing a traffic simulation model and inputting different combinations of adjustable modes into the model, the system simulates future traffic flow within the time zone under these control measures, including changes in vehicle speeds, queue lengths, and travel times. Based on the simulation results, the system evaluates the actual effectiveness of different adjustable mode combinations in reducing congestion on the target regulated road section. The system also considers factors such as implementation costs (such as equipment modification and labor costs) and the impact on surrounding traffic (such as whether it will trigger new congestion points) to select the optimal adjustable mode combination as the traffic flow control strategy.

[0047] Furthermore, the control priority of the signal light timing optimization mode is greater than the control priority of the route guidance mode, which is greater than the control priority of the lane control mode; when the adjustable modes include multiple modes, the mode sequence is optimized first according to the control priority until the control adaptability meets the preset target.

[0048] Preferably, the control priority of the signal timing optimization mode is higher than the control priority of the route guidance mode, which in turn is higher than the control priority of the lane control mode. When the target regulated road section includes multiple adjustable modes, the system will optimize the mode sequence according to the above priorities. Specifically, the system will first adjust the signal timing optimization mode to ensure maximum signal timing efficiency while reducing congestion and delays. Then, the system will optimize the route guidance mode to guide some vehicles to alternative routes, further reducing the burden on the target road section. Finally, if congestion still exists, the system will activate the lane control mode to adjust lane openness or restrict specific vehicle types. The optimization process will continue until the control adaptability meets the preset target. That is, when the traffic flow on the target road section is effectively controlled and the congestion level is reduced to a predetermined threshold, the control strategy is considered complete and the system will cease further control operations.

[0049] In summary, the embodiments of the present application have at least the following technical effects:

[0050] Multi-source data collection is performed through the data acquisition interface to obtain a multi-source traffic dataset. Next, the multi-source traffic dataset is spatially fused to obtain a traffic flow marker map. The current traffic status is then analyzed based on the traffic flow marker map, identifying local and global congestion and constructing a traffic situation map. Finally, based on the traffic situation map, traffic flow status predictions are performed within a preset future time zone. Traffic control decisions are made based on the prediction results, and a traffic flow control strategy is generated. This solves the technical problem of inaccurate traffic flow prediction and control in existing technologies, achieving the technical effect of intelligent dynamic traffic flow control through multi-source data fusion and spatiotemporal analysis, improving traffic management accuracy and response speed.

[0051] Example 2, based on the same inventive concept as the intelligent traffic flow dynamic control method of multi-source data fusion in the above embodiment, Figure 2 As shown, the present application provides an intelligent traffic flow dynamic control system based on multi-source data fusion, wherein the system includes:

[0052] Data acquisition module 11: performs multi-source data acquisition through the data acquisition interface to obtain a multi-source traffic data set; spatial fusion module 12: performs spatial fusion on the multi-source traffic data set to obtain a traffic mark map; analysis module 13: performs current traffic status analysis based on the traffic mark map, identifies local and global congestion, and constructs a traffic situation map; control module 14: performs traffic flow status prediction in a preset future time zone based on the traffic situation map, makes control decisions based on the prediction results, and generates a traffic flow control strategy.

[0053] Furthermore, the data acquisition module 11 is used to perform the following method:

[0054] The data acquisition interface is used to uniformly access multiple types of data sources, which include at least various types of road perception devices, vehicle-mounted systems, and mobile terminals.

[0055] Furthermore, the spatial fusion module 12 is configured to perform the following method:

[0056] The multi-source traffic datasets are aligned to the same time scale to generate an aligned dataset; a road network model is constructed, the aligned datasets are mapped to road links and intersections, and data confidence-weighted fusion is performed to obtain a fusion result; traffic flow features are extracted and marked using the fusion result, including vehicle volume, average speed, and congestion index, to generate the traffic marked map.

[0057] Furthermore, the spatial fusion module 12 is configured to perform the following method:

[0058] The road network model is a graph structure model of a preset road area, wherein nodes are intersections, cross points and preset landmarks; and edges are directed edges representing roads.

[0059] Furthermore, the spatial fusion module 12 is configured to perform the following method:

[0060] Initializing the reliability weights of multiple types of data sources connected to the data acquisition interface to generate initialization weights; performing GPS drift processing on the aligned data set, and then performing weighted fusion on the multi-source data mapped to the road links and intersections using the initialization weights to generate the fusion result.

[0061] Furthermore, the control module 14 is configured to execute the following method:

[0062] The target regulated road section is located according to the prediction result, where the target regulated road section is a road section where the congestion level is greater than a preset congestion threshold and the congestion duration is greater than a preset time threshold; the adjustable modes of the target regulated road section are collected, including one or more of a signal light timing optimization mode, a route induction mode, and a lane control mode; based on the adjustable modes, with the goal of reducing the congestion level, a flow control analysis of the target regulated road section is performed to generate the traffic flow control strategy.

[0063] Furthermore, the control module 14 is configured to execute the following method:

[0064] The control priority of the signal light timing optimization mode is greater than the control priority of the route guidance mode, which is greater than the control priority of the lane control mode; when the adjustable modes include multiple modes, the mode sequence is optimized first according to the control priority until the control adaptability meets the preset target.

[0065] It should be noted that the order in which the embodiments of the present application are presented is for illustrative purposes only and does not necessarily represent the superiority or inferiority of the embodiments. Furthermore, the foregoing descriptions of specific embodiments of this specification are provided. The processes depicted in the accompanying drawings do not necessarily require the specific order or sequential order shown to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0066] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present application shall be included in the scope of protection of the present application.

[0067] This specification and drawings are merely illustrative of the present application and are intended to cover any and all modifications, variations, combinations, or equivalents within the scope of the present application. Obviously, those skilled in the art may make various modifications and variations to the present application without departing from the scope of the present application. Thus, to the extent such modifications and variations fall within the scope of the present application and its equivalents, the present application is intended to include such modifications and variations.

Claims

1. An intelligent traffic flow dynamic control method based on multi-source data fusion, characterized by: The method comprises: Perform multi-source data collection through the data collection interface to obtain a multi-source traffic data set; Performing spatial fusion on the multi-source traffic dataset to obtain a traffic marking map; Analyze the current traffic status based on the traffic mark map, identify local and global congestion, and build a traffic situation map; Based on the traffic situation map, traffic flow state prediction in a preset future time zone is performed, and control decisions are made based on the prediction results to generate a traffic flow control strategy.

2. The intelligent traffic flow dynamic control method based on multi-source data fusion according to claim 1 is characterized in that: The data acquisition interface is used to uniformly access multiple types of data sources, which include at least various types of road perception devices, vehicle-mounted systems, and mobile terminals.

3. The intelligent traffic flow dynamic control method based on multi-source data fusion according to claim 1 is characterized in that: The multi-source traffic dataset is spatially fused to obtain a traffic mark map, including: Aligning the multi-source traffic datasets to the same time scale to generate an aligned dataset; Constructing a road network model, mapping the aligned dataset to road links and intersections, performing data confidence weighted fusion, and obtaining a fusion result; Traffic flow features are extracted and marked based on the fusion results, including vehicle volume, average vehicle speed, and congestion index, to generate the traffic marked map.

4. The intelligent traffic flow dynamic control method based on multi-source data fusion according to claim 3 is characterized in that: The road network model is a graph structure model of a preset road area, wherein nodes are intersections, cross points and preset landmarks; and edges are directed edges representing roads.

5. The intelligent traffic flow dynamic control method based on multi-source data fusion according to claim 4 is characterized in that: Map the aligned dataset to road links and intersections, perform data confidence weighted fusion, and obtain fusion results, including: Initializing the reliability weights of multiple types of data sources connected to the data acquisition interface to generate initialization weights; After performing GPS drift processing on the aligned data set, weighted fusion is performed on the multi-source data mapped to the road links and intersections using the initialization weights to generate the fusion result.

6. The intelligent traffic flow dynamic control method based on multi-source data fusion according to claim 1 is characterized in that: Based on the traffic situation map, traffic flow state prediction is performed in a preset future time zone, and control decisions are made based on the prediction results to generate a traffic flow control strategy, including: Locating a target regulated road section according to the prediction result, wherein the target regulated road section is a road section where the congestion level is greater than a preset congestion threshold and the congestion duration is greater than a preset time threshold; Collecting adjustable modes of the target regulated road section, including one or more of a signal light timing optimization mode, a route guidance mode, and a lane control mode; Based on the adjustable mode, with the goal of reducing congestion, traffic flow control analysis of the target control section is performed to generate the traffic flow control strategy.

7. The intelligent traffic flow dynamic control method based on multi-source data fusion according to claim 6 is characterized in that: The control priority of the signal light timing optimization mode is higher than the control priority of the route guidance mode, which is higher than the control priority of the lane control mode; When the adjustable mode includes multiple modes, the mode sequence is first optimized according to the control priority until the control adaptability meets the preset target.

8. Intelligent traffic flow dynamic control system based on multi-source data fusion, characterized by: The intelligent traffic flow dynamic control method for implementing the multi-source data fusion according to any one of claims 1 to 7, the system comprising: Data acquisition module: performs multi-source data acquisition through the data acquisition interface to obtain multi-source traffic data sets; Spatial fusion module: performs spatial fusion on the multi-source traffic dataset to obtain a traffic mark map; Analysis module: Analyze the current traffic status based on the traffic mark map, identify local and global congestion, and build a traffic situation map; Control module: Based on the traffic situation map, it performs traffic flow state prediction in a preset future time zone, makes control decisions based on the prediction results, and generates a traffic flow control strategy.

Images