Smart city intelligent traffic management system based on big data

Abstract

The invention relates to the technical field of traffic management, in particular to a smart city intelligent traffic management system based on big data, which comprises a multi-source data acquisition module, a data fusion module, a dynamic feature generation module, a situation prediction module, a multi-target optimization module and an instruction execution module, the multi-source data acquisition module acquires multi-source data in real time; the data fusion module generates a standardized traffic data flow and outputs a three-dimensional data cube; the dynamic feature generation module outputs a traffic feature matrix; the situation prediction module generates a dynamic situation map; the multi-objective optimization module outputs an optimization instruction set; and the instruction execution module issues a control instruction to the traffic signal controller, the variable information board and the vehicle-mounted terminal. According to the invention, traffic jam can be relieved, road traffic efficiency can be improved, bus priority scheduling can be optimized, vehicle delay can be reduced, and emergency response speed can be improved.

Description

Technical Field

[0001] The present invention relates to the field of traffic management technology, and in particular to a smart city intelligent traffic management system based on big data. Background Art

[0002] With the acceleration of urbanization, the number of motor vehicles continues to grow. Urban traffic systems are facing problems such as road congestion, delayed signal control, and slow accident response, which seriously affect residents' travel efficiency and traffic safety. In order to alleviate urban traffic pressure and improve road utilization, countries have promoted the construction of smart transportation and used big data, artificial intelligence, the Internet of Things (IoT) and V2X communication technologies to perceive, analyze and optimize traffic flows in real time. However, traditional traffic management systems still rely on fixed-cycle signal control, static path planning and other methods, which are difficult to meet the complex and changeable modern urban traffic needs. A more intelligent and precise traffic optimization system is urgently needed to achieve efficient traffic scheduling and management.

[0003] The current intelligent traffic management system has many shortcomings in data fusion, traffic situation prediction and optimization control. First, the singleness of data collection. Most systems only rely on traffic cameras or geomagnetic sensors, and lack the integration of multi-source data such as GPS positioning and meteorological information, resulting in incomplete data and unable to accurately reflect the real-time road conditions. Second, the data processing capacity is limited. The existing system lacks time-space benchmark alignment and semantic mapping conversion mechanism, making it difficult to effectively integrate heterogeneous data, affecting the accuracy of traffic situation analysis. Third, the prediction and optimization algorithms are insufficient. Traditional methods often rely on historical statistical data and cannot accurately predict short-term traffic flow changes, resulting in signal control and path scheduling lags. Fourth, there is a lack of intelligent scheduling mechanism. Most of the existing traffic light control systems are based on fixed schedules and fail to make full use of real-time traffic data and dynamic green wave optimization, resulting in uneven traffic flow distribution and increased congestion. In addition, the existing system does not take into account the priority scheduling of buses and the passage needs of emergency vehicles, affecting public transportation efficiency and emergency response speed.

[0004] In response to the above problems, the present invention provides a smart city intelligent traffic management system based on big data, which can effectively improve road traffic efficiency, reduce traffic congestion, optimize public transportation scheduling, and improve the overall level of urban intelligent traffic management. Summary of the invention

[0005] The present invention provides a smart city intelligent traffic management system based on big data.

[0006] The smart city intelligent traffic management system based on big data includes a multi-source data acquisition module, a data fusion module, a dynamic feature generation module, a situation prediction module, a multi-objective optimization module and an instruction execution module, among which; The multi-source data acquisition module acquires multi-source data in real time, including traffic camera video streams, geomagnetic sensor pulse signals, vehicle-mounted GPS positioning data and meteorological information; The data fusion module generates a standardized traffic data stream by performing spatiotemporal benchmark alignment and semantic mapping conversion on multi-source data, and outputs a three-dimensional data cube including timestamps, spatial coordinates, and event types; The dynamic feature generation module extracts the spatiotemporal features of traffic flow density, average vehicle speed, and road occupancy rate based on the output three-dimensional data cube, and outputs a traffic feature matrix; The situation prediction module generates a dynamic situation map including traffic flow distribution and accident probability prediction in the next 15 minutes based on the output traffic feature matrix and the spatiotemporal graph convolution network; The multi-objective optimization module generates an optimal path allocation scheme based on the dynamic situation map, combined with the bus priority coefficient and the emergency vehicle traffic demand, and outputs an optimization instruction set including lane-level control instructions and signal light timing parameters; The instruction execution module receives the optimized instruction set and sends control instructions to the traffic signal controller, variable information board, and vehicle-mounted terminal through the V2X communication protocol.

[0007] Optionally, the multi-source data acquisition module includes: Video stream acquisition: Receive the video stream of the traffic camera in real time through the video acquisition terminal, and use the edge computing device to detect and track the target, extract the vehicle position, speed, lane occupancy, and encode it into time series data; Pulse signal acquisition: Pulse signals are collected through geomagnetic sensors, and the number of vehicles passing, parking time, and lane occupancy are calculated based on the signal pulse frequency; Vehicle GPS positioning data acquisition: Receive vehicle positioning information from the vehicle GPS device and perform trajectory smoothing in combination with inertial navigation data to improve positioning accuracy and generate vehicle driving data including timestamp, speed, and direction; Meteorological information acquisition: Receive meteorological information provided by meteorological monitoring stations or meteorological APIs, including rainfall, haze, and road slipperiness, and mark abnormal weather events.

[0008] Optionally, the data fusion module includes: Time-space benchmark alignment: Time synchronization and spatial coordinate standardization of the collected multi-source data, using linear interpolation time alignment method and unified coordinate projection transformation; Semantic mapping conversion: Map multi-source data to a unified set of traffic event types, including vehicle travel V, parking S, congestion C, and accident A, and achieve automatic classification through the Bayesian classification model; Build a standardized traffic data stream: Build a standardized traffic data stream based on aligned multi-source data ; Output three-dimensional data cube: Construct a three-dimensional data cube based on traffic time, spatial location, and event type, and organize data in time dimension, spatial dimension, and event type dimension.

[0009] Optionally, the dynamic feature generation module includes: Traffic flow density calculation: Traffic flow density Indicates the number of vehicles per unit road length and counts the distribution of vehicles based on the traffic event types in the three-dimensional data cube; Vehicle average speed calculation: Vehicle average speed Measures the overall speed level of vehicles traveling on a road section; Road occupancy rate calculation: Road occupancy rate A measure of the extent to which road space is occupied by vehicles; Generate traffic characteristic matrix: Construct traffic characteristic matrix based on the calculation results of traffic flow density, average vehicle speed and road occupancy rate , and integrate multiple spatiotemporal features into structured data.

[0010] Optionally, the situation prediction module includes: Building a traffic space-time graph model: Representing the urban road network as a space-time graph ,in, is a set of nodes in the road network, representing different road sections or intersections. is an edge set, which represents the road topological connection relationship. is the adjacency matrix, which represents the connection strength between road segments and is defined as a Gaussian kernel weight matrix based on geographic distance. ; Time Series Modeling: Using History Traffic characteristic matrix of time steps As input, each matrix includes traffic flow density, average speed, road occupancy rate, and constructs a time series ; Spatiotemporal graph convolution computation: Spatiotemporal graph convolution network (ST-GCN) is used for feature extraction, including spatial graph convolution and temporal convolution, where; The spatial graph convolution is based on the road topology structure and uses the normalized Laplacian matrix Perform graph convolution operations to calculate the high-order feature representation of each node; The temporal convolution uses a one-dimensional causal convolution to extract temporal features; Future traffic flow prediction: The features extracted by the last layer of spatiotemporal graph convolutional network (ST-GCN) The data is passed to the fully connected layer to generate traffic flow forecasts for each road section in the next 15 minutes. Accident probability prediction: Based on the predicted traffic data, the Softmax classification model is used to calculate the accident probability of each road section; Output dynamic situation map: construct dynamic situation map based on the prediction results of future traffic flow and accident probability .

[0011] Optionally, the multi-objective optimization module includes: Traffic flow optimization: Based on the dynamic situation map and combined with the bus priority coefficient, the multi-objective path optimization algorithm is used to dynamically adjust the road traffic strategy to achieve a balance between bus priority scheduling and overall traffic efficiency; Traffic light optimization: Combine dynamic situation maps with emergency vehicle traffic requirements and use an adaptive green wave control algorithm to optimize traffic lights; Optimization instruction set generation: Convert the results of traffic flow optimization and signal light optimization into optimization instruction sets.

[0012] Optionally, the traffic flow optimization includes: Constructing a road traffic network: Constructing a weighted road traffic network based on a dynamic situation map ,in, is the road section weight matrix, which represents the travel cost of each road section; Calculate bus priority path weights: To ensure that buses have priority, adjust the path weights of bus routes ; Calculate the optimal travel path: Use the Dijkstra algorithm to balance the overall traffic flow and calculate the optimal path while taking into account bus priority scheduling; Dynamically adjust road traffic strategies: Based on the calculation results of the optimal traffic path, generate real-time traffic flow optimization plans and implement road traffic strategies, including adjusting bus lanes, dynamic speed limits, adjusting variable lanes, and guiding ordinary vehicles to avoid bus priority roads.

[0013] Optionally, the traffic light optimization includes: Constructing signal control optimization objective function: weighted road traffic network based on dynamic situation map , construct the signal control optimization objective function, the optimization objectives of signal control include maximizing traffic flow efficiency and minimizing the waiting time of emergency vehicles; Detect emergency vehicles and adjust signal priorities: Get the real-time location, speed, and estimated arrival time of emergency vehicles, and calculate the priority of each signal light on its path ; Adaptive green wave control optimization: Traffic light optimization uses an adaptive green wave control algorithm to dynamically adjust the signal light phases of adjacent intersections to optimize vehicle passing efficiency; Optimize traffic light timing and issue instructions: Generate traffic light optimization plans based on the results of adaptive green wave control optimization, including real-time adjustment of green light duration, dynamic adjustment of signal phase difference, and priority release for emergency vehicles.

[0014] Optionally, the adaptive green wave control optimization includes: Calculate the optimal phase difference: Calculate the optimal green wave signal phase difference based on the road length and average vehicle speed ; Green light time adaptive adjustment: for normal traffic flow, green light time Adaptively adjust according to real-time traffic flow.

[0015] Optionally, the optimization instruction set generation includes: Receive optimization results and analyze control parameters: Obtain optimized traffic paths, road traffic strategies, and traffic light optimization solutions from traffic flow optimization and signal light optimization, and analyze control parameters, including optimal driving paths (adjusting lane traffic rules and optimizing traffic flow), signal light timing parameters (green light duration, phase difference), and lane-level control instructions (activation of variable lanes, speed limit adjustment); Optimization instruction formatting and instruction generation: Based on the parsed control parameters, a formatted optimization instruction set is generated, including path optimization instructions (guiding public transportation priority and reducing traffic in congested areas), signal light optimization instructions (adjusting intersection signal light timing to form green wave control), and dynamic traffic scheduling instructions (speed limit adjustment, variable lane switching, and traffic control strategies); Real-time delivery of optimized instruction sets: Through the V2X communication protocol, the optimized instruction sets are delivered in real time to the traffic signal controller (to perform traffic light optimization), the variable message board (VMS, providing route suggestions to drivers), and the in-vehicle terminal (IVI, navigation guidance, prompting the optimal route for buses and social vehicles).

[0016] Beneficial effects of the present invention: The present invention realizes all-round monitoring, analysis and optimization of urban traffic through multi-source data collection, data fusion, dynamic feature extraction, situation prediction, multi-objective optimization and instruction execution modules. The system integrates video streams, geomagnetic sensors, GPS and meteorological information to ensure the comprehensiveness and high accuracy of data. It uses spatiotemporal benchmark alignment and semantic mapping conversion to construct a three-dimensional data cube to improve data consistency and availability. It extracts key indicators such as traffic flow density, average vehicle speed, and road occupancy rate through dynamic feature generation, providing accurate data support for traffic status analysis. The present invention, through the situation prediction module of the spatiotemporal graph convolutional network, can accurately predict the traffic flow changes and accident probability in the next 15 minutes, providing a scientific basis for optimized scheduling. The multi-objective optimization module combines public transportation priority, emergency vehicle demand and real-time traffic status to optimize the optimal path allocation, signal light control and lane-level scheduling strategy. The signal light timing is dynamically adjusted through the adaptive green wave control algorithm to ensure the rapid passage of emergency vehicles while maintaining the balance of the overall traffic flow. The optimization instruction set generation module converts the optimization results into standardized instructions and issues them in real time through the V2X communication protocol to achieve accurate scheduling and automatic execution of intelligent transportation. The present invention improves the intelligence, precision and real-time nature of traffic management through multi-source data fusion, intelligent prediction and optimized control, can effectively alleviate traffic congestion, improve road traffic efficiency, optimize bus priority scheduling, reduce vehicle delays, and increase emergency response speed. The system has high adaptability and scalability and can be widely used in traffic management of smart cities to enhance the comprehensive traffic scheduling capabilities of cities. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only for the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0018] Figure 1 A schematic diagram of system function modules according to an embodiment of the present invention; Figure 2 Schematic diagram of a situation prediction module according to an embodiment of the present invention. DETAILED DESCRIPTION

[0019] The present invention is described in detail below in conjunction with the accompanying drawings and specific embodiments. At the same time, it is explained here that in order to make the embodiments more detailed, the following embodiments are the best and preferred embodiments, and those skilled in the art may also adopt other alternatives to implement some known technologies; and the accompanying drawings are only for more specific description of the embodiments, and are not intended to specifically limit the present invention.

[0020] It should be noted that the references to "one embodiment", "embodiment", "exemplary embodiments", "some embodiments" and the like in the specification indicate that the embodiments described may include specific features, structures or characteristics, but not every embodiment may include the specific features, structures or characteristics. In addition, when a specific feature, structure or characteristic is described in conjunction with an embodiment, it should be within the knowledge of a person skilled in the art to implement such feature, structure or characteristic in conjunction with other embodiments (whether or not explicitly described).

[0021] In general, a term can be understood, at least in part, from its use in context. For example, depending, at least in part, on the context, the term "one or more" as used herein can be used to describe any feature, structure, or characteristic in the singular sense, or can be used to describe a combination of features, structures, or characteristics in the plural sense. Additionally, the term "based on" can be understood as not necessarily intended to convey an exclusive set of factors, but can instead, depending, at least in part, on the context, allow for the presence of other factors that are not necessarily explicitly described.

[0022] like Figure 1-Figure 2 As shown, the smart city intelligent traffic management system based on big data includes a multi-source data acquisition module, a data fusion module, a dynamic feature generation module, a situation prediction module, a multi-objective optimization module and an instruction execution module, wherein; The multi-source data acquisition module acquires multi-source data in real time, including traffic camera video streams, geomagnetic sensor pulse signals, vehicle-mounted GPS positioning data and meteorological information; The data fusion module generates standardized traffic data streams by performing spatiotemporal benchmark alignment and semantic mapping conversion on multi-source data, and outputs a three-dimensional data cube including timestamps, spatial coordinates, and event types; The dynamic feature generation module extracts spatiotemporal features such as traffic flow density, average vehicle speed, and road occupancy rate based on the output three-dimensional data cube, and outputs a traffic feature matrix; The situation prediction module generates a dynamic situation map including traffic flow distribution and accident probability prediction in the next 15 minutes based on the output traffic feature matrix and the spatiotemporal graph convolution network; The multi-objective optimization module generates the optimal path allocation plan based on the dynamic situation map, combined with the bus priority coefficient and the emergency vehicle traffic demand, and outputs the optimization instruction set including lane-level control instructions and signal light timing parameters; The command execution module receives the optimized command set and sends control commands to the traffic signal controller, variable information board, and vehicle terminal through the V2X communication protocol.

[0023] The multi-source data acquisition module includes: Video stream acquisition: Receive the video stream of the traffic camera in real time through the video acquisition terminal, and use the edge computing device to detect and track the target, extract the vehicle position, speed, lane occupancy, and encode it into time series data; Pulse signal acquisition: Pulse signals are collected through geomagnetic sensors, and the number of vehicles passing, parking time, and lane occupancy are calculated based on the signal pulse frequency; Vehicle GPS positioning data acquisition: Receive vehicle positioning information from the vehicle GPS device and perform trajectory smoothing in combination with inertial navigation data to improve positioning accuracy and generate vehicle driving data including timestamp, speed, and direction; Meteorological information acquisition: Receive meteorological information provided by meteorological monitoring stations or meteorological APIs, including rainfall, haze, and road slipperiness, and mark abnormal weather events; Through the above content, video streams, geomagnetic sensors, GPS and meteorological information are integrated to ensure the comprehensiveness and high accuracy of traffic data. Video streams monitor vehicle dynamics in real time, geomagnetic sensors accurately calculate vehicle flow and parking time, GPS combined with inertial navigation improves positioning accuracy, and meteorological information identifies the impact of severe weather. This builds an accurate and real-time traffic data foundation, providing efficient support for intelligent traffic management.

[0024] The data fusion module includes: Time-space benchmark alignment: Time synchronization and spatial coordinate standardization of the collected multi-source data, using linear interpolation time alignment method and unified coordinate projection transformation, specifically including: (1) Time synchronization processing: Assume that the timestamp set of each sensor data is , the base timestamp set is ,in is the original timestamp, In order to unify the time scale of the alignment, linear interpolation is used to calculate the unaligned data, expressed as: ; in, is the data value after alignment, and are adjacent original data values, , For The closest timestamp; (2) Spatial coordinate conversion: All location information is uniformly converted to the universal geographic coordinate system (WGS-84). Perform projection transformation, expressed as: ; ; ; in, , are longitude and latitude respectively, is the radius of the Earth, , , is the transformed geocentric coordinate; Semantic mapping conversion: Map multi-source data to a unified set of traffic event types, including vehicle travel V, parking S, congestion C, and accident A, and realize automatic classification through the Bayesian classification model, which is expressed as: ; in, For a given feature Traffic incident types The probability of Traffic incident type Characteristics when it occurs The conditional probability of For event type The prior probability of Traffic incident type Characteristics when it occurs The conditional probability of For event type The prior probability of Build a standardized traffic data stream: Build a standardized traffic data stream based on aligned multi-source data , expressed as: ; in, To unify the timestamp, is the corresponding traffic event type; Output three-dimensional data cube: Construct a three-dimensional data cube based on traffic time, spatial location, and event type, and organize data in time dimension, space dimension, and event type dimension, expressed as: ; in, For a three-dimensional data cube in time ,space , Event Type The value at To standardize traffic data flow, is the data weight, is the Kronecker Delta function, which is used to indicate whether the data belongs to this category. ; Through the above content, accurate alignment of data across devices can be achieved, time drift and spatial errors can be eliminated, the quality of data fusion can be improved, and the Bayesian classification model can be used to perform intelligent semantic mapping of data from different sources, automatically identify traffic event types, and enhance the system's understanding ability. The final three-dimensional data cube constructed integrates time, space and event information.

[0025] The dynamic feature generation module includes: Traffic flow density calculation: Traffic flow density Represents the number of vehicles per unit road length, and the distribution of vehicles is statistically analyzed based on the traffic event types in the three-dimensional data cube, expressed as: ; in, For road section The traffic flow density at For time window The total number of vehicles on the road segment. is the length of the road section; Vehicle average speed calculation: Vehicle average speed Measures the overall speed level of vehicles on the road section, expressed as: ; in, For road section The average speed of vehicles at For the The speed of the vehicle; Road occupancy rate calculation: Road occupancy rate Measures the degree to which road space is occupied by vehicles, expressed as: ; in, For road section The road occupancy rate at For the Average length of a vehicle , is the length of the road section; Generate traffic characteristic matrix: Construct traffic characteristic matrix based on the calculation results of traffic flow density, average vehicle speed and road occupancy rate , and integrate multiple spatiotemporal features into structured data, expressed as: ; The row index Represents the time window, column index Represents the road location; Through the above content, accurate and structured data support is provided for intelligent traffic management. By real-time calculation of road section traffic density, road congestion can be accurately portrayed, overall traffic efficiency can be evaluated based on the average vehicle speed, signal control strategy can be optimized, and road occupancy rate can be used to measure road resource usage. Dynamic scheduling and lane optimization can be assisted to ensure the efficiency and real-time nature of data processing, provide a solid data foundation for subsequent traffic situation prediction and optimization decision-making, and improve the accuracy and response speed of the intelligent transportation system.

[0026] The situation prediction module includes: Building a traffic space-time graph model: Representing the urban road network as a space-time graph ,in, is a set of nodes in the road network, representing different road sections or intersections. is an edge set, which represents the road topological connection relationship. is the adjacency matrix, which represents the connection strength between road segments and is defined as a Gaussian kernel weight matrix based on geographic distance. ; ; in, For Node and The geographical distance between is the scale parameter of the Gaussian kernel, which is used to control the decay speed; Time Series Modeling: Using History Traffic characteristic matrix of time steps As input, each matrix includes traffic flow density, average speed, road occupancy rate, and constructs a time series , expressed as: ; in, , is the number of nodes (number of road sections), is the feature dimension, is the time window length; Spatiotemporal graph convolution computation: Spatiotemporal graph convolution network (ST-GCN) is used for feature extraction, including spatial graph convolution and temporal convolution, where; Spatial graph convolution is based on road topology and uses a normalized Laplacian matrix Perform graph convolution operations and calculate the high-order feature representation of each node, expressed as: ; in, For the The feature matrix of the layer, For the The feature matrix of the layer, are trainable weights, is the normalized Laplace matrix, is the degree matrix, is the ReLU activation function; Temporal convolution uses one-dimensional causal convolution to extract temporal features, expressed as: ; in, is the temporal convolution kernel weight, is the bias term, is the feature matrix after time convolution; Future traffic flow prediction: The features extracted by the last layer of spatiotemporal graph convolutional network (ST-GCN) The data is passed to the fully connected layer to generate the traffic flow forecast for each road section in the next 15 minutes, expressed as: ; in, is the predicted traffic flow matrix, and is the trainable parameter of the fully connected layer; Accident probability prediction: Based on the predicted traffic data, the Softmax classification model is used to calculate the accident probability of each road section, which is expressed as: ; in, For road section The probability of an accident occurring, , are the weights and biases of the classification model, respectively. , All possible accident categories The corresponding weights and bias terms; Output dynamic situation map: construct dynamic situation map based on the prediction results of future traffic flow and accident probability , expressed as: ; in, is the probability of accident occurrence; Through the above content, the traffic flow distribution and accident probability in the next 15 minutes can be accurately predicted, and a dynamic situation map can be generated. The relationship between roads can be modeled through spatial graph convolution, and the time convolution is used to capture the traffic time series changes to ensure that the prediction results are both global and dynamic. The road section weights are optimized through the Gaussian kernel adjacency matrix to improve the accuracy of traffic flow prediction. The accident probability is calculated based on the Softmax classification model to achieve accurate early warning. Finally, the generated dynamic situation map can provide real-time decision support for the traffic management system, improve the intelligence level of signal optimization, road scheduling and accident warning, and help smart cities operate efficiently.

[0027] The multi-objective optimization module includes: Traffic flow optimization: Based on the dynamic situation map and combined with the bus priority coefficient, the multi-objective path optimization algorithm is used to dynamically adjust the road traffic strategy to achieve a balance between bus priority scheduling and overall traffic efficiency; Traffic light optimization: Combining dynamic situation maps with emergency vehicle traffic needs, an adaptive green wave control algorithm is used to optimize traffic lights to ensure that emergency vehicles (such as ambulances and fire trucks) have priority while maintaining the balance of overall traffic flow; Optimization instruction set generation: convert the results of traffic flow optimization and signal light optimization into optimization instruction sets; Through the above content, intelligent management of urban traffic is realized. Traffic flow optimization uses dynamic situation maps and multi-objective path optimization algorithms to ensure priority scheduling of buses while balancing the overall traffic flow and improving road traffic efficiency. Traffic light optimization is based on an adaptive green wave control algorithm, which gives priority to the rapid passage of emergency vehicles and optimizes traffic light timing to reduce traffic congestion and realize intelligent and dynamic traffic control, which can significantly improve road traffic efficiency, reduce traffic delays, optimize public transportation services, and increase emergency response speed.

[0028] Traffic flow optimization includes: Constructing a road traffic network: Constructing a weighted road traffic network based on a dynamic situation map ,in, is the road section weight matrix, which represents the travel cost of each road section; Road segment weight matrix It is expressed as: ; in, For road section arrive The travel cost, is the estimated travel time of the road section under the current traffic flow status, The degree of road congestion, is the accident risk factor, is the bus priority coefficient, , , , is the weight coefficient; Calculate bus priority path weights: To ensure that buses have priority, adjust the path weights of bus routes , expressed as: ; in, The travel cost of the optimized road section for public transportation, is the bus priority dispatch coefficient. The higher the value, the more buses tend to choose low-priority roads to balance the overall traffic flow. Calculate the optimal travel path: The Dijkstra algorithm is used to balance the overall traffic flow while considering bus priority scheduling, and calculate the optimal path, which is expressed as: ; in, For path The total travel cost, is the bus route from the starting point to the end point, are adjacent segments in the path; Dynamically adjust road traffic strategies: Based on the calculation results of the optimal traffic path, generate real-time traffic flow optimization solutions and implement road traffic strategies, including adjusting bus lanes, dynamic speed limits, adjusting variable lanes, and guiding ordinary vehicles to avoid bus priority roads; The road access strategy includes: Adjust bus lanes: In high-congestion areas, open some private vehicles to traffic in due course to improve road utilization; Dynamic speed limit: lower or raise the speed limit on some sections of the road to divert traffic to non-congested areas; Adjust variable lanes: flexibly change the direction of variable lanes according to traffic conditions to improve traffic capacity; Directing ordinary vehicles to avoid bus priority roads: reducing non-bus occupancy on bus routes; Through the above content, based on the dynamic situation map, combined with the bus priority coefficient, the multi-objective path optimization algorithm is used to dynamically adjust the road traffic strategy to achieve a balance between bus priority and overall traffic flow. By constructing a weighted road traffic network, the system can intelligently optimize the traffic paths of buses and social vehicles, reduce congestion, and improve traffic efficiency. At the same time, combined with dynamic speed limits, variable lane adjustments and bus lane optimization, it can flexibly adapt to real-time traffic changes and improve road resource utilization.

[0029] Traffic light optimization includes: Constructing signal control optimization objective function: weighted road traffic network based on dynamic situation map , construct the signal control optimization objective function. The optimization objectives of signal control include maximizing traffic flow efficiency and minimizing the waiting time of emergency vehicles. The optimization objective function is expressed as: ; in, is the total number of intersections, For the The average waiting time of vehicles at each intersection, is the traffic flow density at the intersection, is the emergency vehicle priority demand coefficient at the intersection, , , is the weight coefficient; Detect emergency vehicles and adjust signal priorities: Get the real-time location, speed, and estimated arrival time of emergency vehicles, and calculate the priority of each signal light on its path , expressed as: ; in, For intersection The priority of emergency vehicles at The estimated arrival time of the emergency vehicle. is the current time, To adjust the parameters; Adaptive green wave control optimization: Traffic light optimization uses an adaptive green wave control algorithm to dynamically adjust the signal light phases of adjacent intersections to optimize vehicle passing efficiency; Optimize the traffic light timing plan and issue instructions: Based on the results of adaptive green wave control optimization, generate traffic light optimization plans, including real-time adjustment of green light duration, dynamic adjustment of signal phase difference, and priority release for emergency vehicles; Through the above contents, the road traffic efficiency is improved and the priority passage of emergency vehicles is guaranteed. The green light duration can be dynamically optimized according to the real-time traffic flow, the vehicle waiting time can be reduced, and the traffic efficiency of adjacent intersections can be improved through green wave coordinated control. At the same time, the system can accurately predict the arrival time of emergency vehicles, automatically adjust the traffic lights to give priority, and ensure the rapid passage of emergency vehicles such as ambulances and fire trucks. The optimized traffic light control plan can be sent to the traffic management system in real time, which improves the flexibility of traffic scheduling, alleviates urban congestion, and enhances the intelligence level of smart transportation.

[0030] Adaptive green wave control optimization includes: Calculate the optimal phase difference: Calculate the optimal green wave signal phase difference based on the road length and average vehicle speed , expressed as: ; in, For intersection and The green light phase difference between : Intersection and The length of the road between is the average driving speed of the road section, is the signal period; Green light time adaptive adjustment: for normal traffic flow, green light time Adaptive adjustment based on real-time traffic flow is expressed as: ; in, For the current intersection The green light duration, is the minimum green light duration, is the adjustment factor, is the real-time traffic density of the current lane, is the preset flow threshold; Preset traffic threshold The settings are based on historical data, including: (1) The system collects historical traffic flow data of different time periods and different road types, and calculates the average flow density of the road section, which is expressed as: ; in, is the historical average traffic density of the road section, is the total amount of data within the statistical time period, For the The observed flow density; (2) Calculate the maximum traffic capacity of the road: Calculate the maximum carrying capacity (saturated flow) of the road section according to the road design standards. , which is determined by the road type and the number of lanes, and is expressed as: ; in, is the maximum flow rate that can be carried per unit length of the road section. The maximum traffic capacity per unit time for a single lane, determined by the road design standard, such as highway 2000 vehicles / hour, urban roads Vehicles / hour, is the number of lanes of the road; (3) Set the flow threshold: flow threshold Should be between the historical average flow and maximum capacity So that the system can dynamically adjust the traffic light timing according to the current traffic flow, expressed as: ; in, is the adjustment factor (value range ); Through the above content, vehicles can travel smoothly between multiple traffic lights, reduce parking and waiting, and improve road traffic efficiency. The system optimizes the green wave band according to the road length and average vehicle speed to reduce traffic delays. At the same time, it adaptively adjusts the green light duration based on real-time traffic flow to ensure the smooth flow of high-traffic sections and optimizes signal timing during low-traffic periods to improve overall traffic fluidity. This optimization strategy can effectively alleviate traffic congestion, improve traffic efficiency, and flexibly adapt to sudden traffic changes, realize intelligent and dynamic signal control, and improve the accuracy and intelligence level of urban road management.

[0031] Optimized instruction set generation includes: Receive optimization results and analyze control parameters: Obtain optimized traffic paths, road traffic strategies, and traffic light optimization solutions from traffic flow optimization and signal light optimization, and analyze control parameters, including optimal driving paths (adjusting lane traffic rules and optimizing traffic flow), signal light timing parameters (green light duration, phase difference), and lane-level control instructions (activation of variable lanes, speed limit adjustment); Optimization instruction formatting and instruction generation: Based on the parsed control parameters, a formatted optimization instruction set is generated, including path optimization instructions (guiding public transportation priority and reducing traffic in congested areas), signal light optimization instructions (adjusting intersection signal light timing to form green wave control), and dynamic traffic scheduling instructions (speed limit adjustment, variable lane switching, and traffic control strategies); Real-time delivery of optimized instruction sets: Through the V2X communication protocol, the optimized instruction sets are delivered in real time to the traffic signal controller (to perform signal light optimization), the variable message board (VMS, providing route suggestions to drivers), and the in-vehicle terminal (IVI, navigation guidance, prompting the optimal route for buses and social vehicles); Through the above content, it is ensured that the traffic optimization plan can be executed quickly and efficiently. The system extracts the optimal driving path, signal light timing parameters, and lane-level control instructions from traffic flow optimization and signal light optimization, and converts them into a standardized optimization instruction set to achieve precise scheduling. Through the V2X communication protocol, the optimization instructions can be sent to the traffic signal controller, variable message board (VMS) and vehicle-mounted terminal (IVI) in real time to guide public transportation priority, optimize green wave signals, dynamically adjust lanes and speed limits, ensure that urban road traffic is more efficient, safe and intelligent, improve the real-time response capability of the traffic management system, optimize traffic flow organization, alleviate congestion, and enhance the collaborative scheduling capability of the smart transportation system.

[0032] The present invention covers any substitution, modification, equivalent method and scheme made on the essence and scope of the present invention. In order to make the public have a thorough understanding of the present invention, specific details are described in detail in the following preferred embodiments of the present invention, but those skilled in the art can fully understand the present invention without the description of these details. In addition, in order to avoid unnecessary confusion about the essence of the present invention, well-known methods, processes, procedures, components and circuits are not described in detail.

[0033] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principle of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. The smart city intelligent traffic management system based on big data is characterized by: It includes multi-source data acquisition module, data fusion module, dynamic feature generation module, situation prediction module, multi-objective optimization module and instruction execution module, among which; The multi-source data acquisition module acquires multi-source data in real time, including traffic camera video streams, geomagnetic sensor pulse signals, vehicle-mounted GPS positioning data and meteorological information; The data fusion module generates a standardized traffic data stream by performing spatiotemporal benchmark alignment and semantic mapping conversion on multi-source data, and outputs a three-dimensional data cube including timestamps, spatial coordinates, and event types; The dynamic feature generation module extracts the spatiotemporal features of traffic flow density, average vehicle speed, and road occupancy rate based on the output three-dimensional data cube, and outputs a traffic feature matrix; The situation prediction module generates a dynamic situation map including traffic flow distribution and accident probability prediction in the next 15 minutes based on the output traffic feature matrix and the spatiotemporal graph convolution network; The multi-objective optimization module generates an optimal path allocation scheme based on the dynamic situation map, combined with the bus priority coefficient and the emergency vehicle traffic demand, and outputs an optimization instruction set including lane-level control instructions and signal light timing parameters; The instruction execution module receives the optimized instruction set and sends control instructions to the traffic signal controller, variable information board, and vehicle-mounted terminal through the V2X communication protocol.

2. The smart city intelligent traffic management system based on big data according to claim 1 is characterized in that: The multi-source data acquisition module comprises: Video stream acquisition: Receive the video stream of the traffic camera in real time through the video acquisition terminal, and use the edge computing device to detect and track the target, extract the vehicle position, speed, lane occupancy, and encode it into time series data; Pulse signal acquisition: Pulse signals are collected through geomagnetic sensors, and the number of vehicles passing, parking time, and lane occupancy are calculated based on the signal pulse frequency; Vehicle GPS positioning data acquisition: Receive vehicle positioning information from the vehicle GPS device and perform trajectory smoothing in combination with inertial navigation data to improve positioning accuracy and generate vehicle driving data including timestamp, speed, and direction; Meteorological information acquisition: Receive meteorological information provided by meteorological monitoring stations or meteorological APIs, including rainfall, haze, and road slipperiness, and mark abnormal weather events.

3. The smart city intelligent traffic management system based on big data according to claim 1 is characterized in that: The data fusion module comprises: Time-space benchmark alignment: Time synchronization and spatial coordinate standardization of the collected multi-source data, using linear interpolation time alignment method and unified coordinate projection transformation; Semantic mapping conversion: Map multi-source data to a unified set of traffic event types, including vehicle travel V, parking S, congestion C, and accident A, and achieve automatic classification through the Bayesian classification model; Build a standardized traffic data stream: Build a standardized traffic data stream based on aligned multi-source data ; Output three-dimensional data cube: Construct a three-dimensional data cube based on traffic time, spatial location, and event type, and organize data in time dimension, spatial dimension, and event type dimension.

4. The smart city intelligent traffic management system based on big data according to claim 1 is characterized in that: The dynamic feature generation module comprises: Traffic flow density calculation: Traffic flow density Indicates the number of vehicles per unit road length and counts the distribution of vehicles based on the traffic event types in the three-dimensional data cube; Vehicle average speed calculation: Vehicle average speed Measures the overall speed level of vehicles traveling on a road section; Road occupancy rate calculation: Road occupancy rate A measure of the extent to which road space is occupied by vehicles; Generate traffic characteristic matrix: Construct traffic characteristic matrix based on the calculation results of traffic flow density, average vehicle speed and road occupancy rate , and integrate multiple spatiotemporal features into structured data.

5. The smart city intelligent traffic management system based on big data according to claim 4 is characterized in that: The situation prediction module comprises: Building a traffic space-time graph model: Representing the urban road network as a space-time graph ,in, is a set of nodes in the road network, representing different road sections or intersections. is an edge set, which represents the road topological connection relationship. is the adjacency matrix, which represents the connection strength between road segments and is defined as a Gaussian kernel weight matrix based on geographic distance. ; Time Series Modeling: Using History Traffic characteristic matrix of time steps As input, each matrix includes traffic flow density, average speed, road occupancy rate, and constructs a time series ; Spatiotemporal graph convolution calculation: Spatiotemporal graph convolution network is used for feature extraction, including spatial graph convolution and temporal convolution, where; The spatial graph convolution is based on the road topology structure and uses the normalized Laplacian matrix Perform graph convolution operations to calculate the high-order feature representation of each node; The temporal convolution uses a one-dimensional causal convolution to extract temporal features; Future traffic flow prediction: The features extracted by the last layer of spatiotemporal graph convolutional network The data is passed to the fully connected layer to generate traffic flow forecasts for each road section in the next 15 minutes. Accident probability prediction: Based on the predicted traffic data, the Softmax classification model is used to calculate the accident probability of each road section; Output dynamic situation map: construct dynamic situation map based on the prediction results of future traffic flow and accident probability .

6. The smart city intelligent traffic management system based on big data according to claim 5 is characterized in that: The multi-objective optimization module includes: Traffic flow optimization: Based on the dynamic situation map and combined with the bus priority coefficient, the multi-objective path optimization algorithm is used to dynamically adjust the road traffic strategy to achieve a balance between bus priority scheduling and overall traffic efficiency; Traffic light optimization: Combine dynamic situation maps with emergency vehicle traffic requirements and use an adaptive green wave control algorithm to optimize traffic lights; Optimization instruction set generation: Convert the results of traffic flow optimization and signal light optimization into optimization instruction sets.

7. The smart city intelligent traffic management system based on big data according to claim 6 is characterized in that: The traffic flow optimization includes: Constructing a road traffic network: Constructing a weighted road traffic network based on a dynamic situation map ,in, is the road section weight matrix, which represents the travel cost of each road section; Calculate bus priority path weights: To ensure that buses have priority, adjust the path weights of bus routes ; Calculate the optimal travel path: Use the Dijkstra algorithm to balance the overall traffic flow and calculate the optimal path while taking into account bus priority scheduling; Dynamically adjust road traffic strategies: Based on the calculation results of the optimal traffic path, generate real-time traffic flow optimization plans and implement road traffic strategies, including adjusting bus lanes, dynamic speed limits, adjusting variable lanes, and guiding ordinary vehicles to avoid bus priority roads.

8. The smart city intelligent traffic management system based on big data according to claim 7 is characterized in that: The signal light optimization includes: Constructing signal control optimization objective function: weighted road traffic network based on dynamic situation map , construct the signal control optimization objective function, the optimization objectives of signal control include maximizing traffic flow efficiency and minimizing the waiting time of emergency vehicles; Detect emergency vehicles and adjust signal priorities: Get the real-time location, speed, and estimated arrival time of emergency vehicles, and calculate the priority of each signal light on its path ; Adaptive green wave control optimization: Traffic light optimization uses an adaptive green wave control algorithm to dynamically adjust the signal light phases of adjacent intersections to optimize vehicle passing efficiency; Optimize traffic light timing and issue instructions: Generate traffic light optimization plans based on the results of adaptive green wave control optimization, including real-time adjustment of green light duration, dynamic adjustment of signal phase difference, and priority release for emergency vehicles.

9. The smart city intelligent traffic management system based on big data according to claim 8 is characterized in that: The adaptive green wave control optimization includes: Calculate the optimal phase difference: Calculate the optimal green wave signal phase difference based on the road length and average vehicle speed ; Green light time adaptive adjustment: for normal traffic flow, green light time Adaptively adjust according to real-time traffic flow.

10. The smart city intelligent traffic management system based on big data according to claim 9 is characterized in that: The optimization instruction set generation includes: Receive optimization results and analyze control parameters: Obtain optimized travel paths, road travel strategies, and signal light optimization solutions from traffic flow optimization and signal light optimization, and analyze control parameters, including optimal travel paths, signal light timing parameters, and lane-level control instructions; Optimization instruction formatting and instruction generation: Based on the parsed control parameters, a formatted optimization instruction set is generated, including path optimization instructions, signal light optimization instructions, and dynamic traffic scheduling instructions; Real-time delivery of optimized instruction sets: Through the V2X communication protocol, the optimized instruction sets are delivered to traffic signal controllers, variable information boards, and vehicle-mounted terminals in real time.

Images