Intelligent information release management system and method for urban traffic

Abstract

The invention discloses an intelligent information release management system and method for urban traffic, and relates to the technical field of intelligent traffic management, and the system comprises a database, a dynamic collection module, an urban map module, a demand generation module, a change distribution module and a release uploading module. The technical key points are as follows: real-time traffic road information of urban traffic is collected in real time, a composite network is constructed, a corresponding initial path is generated by analyzing a traffic plan of a traveler and considering multiple factors such as time, a navigation principle and a financial principle, and when the traveler deviates from the initial path, path change is generated, so that the traveler is driven to travel. The influence range generated by each change and the corresponding road section flow on each sub-network are recorded, the road impedance is calculated to serve as a connection function, the composite network is updated in time to provide rich traffic choices, the principle of adjustment according to the real-time traffic condition and the user demand is ensured, and the real-time traffic change is realized to a certain extent. And the management efficiency of the system is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent traffic management, and in particular to a smart information publishing management system and method for urban traffic. Background Art

[0002] With the acceleration of urbanization, the importance of urban public transportation systems in the daily travel destinations of residents has become increasingly prominent. As the two pillars of urban public transportation, the coordinated operation of subways and buses is of great significance to improving overall traffic efficiency and reducing traffic congestion. At the same time, traffic information release refers to the process of providing the public with information about traffic conditions, routes, public transportation, etc. through various channels and technical means. An effective traffic information release system can help travelers understand road conditions, traffic flow, accident information, weather changes, etc. in real time, so as to better plan their travel. Traditional intelligent traffic release management is usually divided into four independent stages: traffic generation, traffic distribution, traffic mode planning and traffic allocation. On the one hand, in the process of traffic generation and traffic distribution, the main analysis is on path planning. At present, the shortest path algorithm is already a basis for traffic navigation. However, the existing intelligent traffic release management system cannot effectively integrate other factors (such as time and finance) for path analysis, and the process lacks accuracy. On the other hand, in the process of traffic mode planning, only a single mode of transportation (such as car) is considered for analysis, and the actual transfer behavior or path change behavior of travelers is often ignored. There is a lack of ability to adjust according to real-time traffic conditions and user needs, which greatly reduces practicality. Furthermore, in traffic allocation, the allocated path plan is not accurate enough to meet the needs of travelers, which reduces travelers' satisfaction and reduces the overall quality and management efficiency of public services. Summary of the invention

[0003] 1. Technical issues to be resolved In view of the deficiencies in the prior art, the present invention provides an intelligent information release management system and method for urban traffic, including a database, a dynamic collection module, a city map module, a demand generation module, a change allocation module and a release upload module. By real-time collection of actual traffic road information of urban traffic, a composite network is constructed, including a first network, a second network and a third network. By analyzing the path changes of travelers, by recording the impact range of each change and the corresponding road traffic on each sub-network, the road impedance is calculated as a connection function, and the composite network is updated in time to provide rich traffic options, thereby solving the problems raised in the background technology.

[0004] (II) Technical solution To achieve the above objectives, the present invention is implemented through the following technical solutions: A smart information publishing and management system for urban transportation, comprising: a database, a dynamic collection module, a city map module, a demand generation module, a change allocation module and a publishing and uploading module; The city map module is used to establish a city mathematical model map based on the OSM map model and the real city map as the background, adding traffic hub intersections, path starting points, and path end points; The dynamic acquisition module is used to collect real-time traffic road information of urban traffic according to the urban mathematical model map, wherein the real-time traffic road information at least includes automobile road information, bus road information and subway road information; The demand generation module is used to establish a composite network according to the real-time traffic road information, input the traveler's traffic plan into the composite network, generate an initial path based on the A* algorithm according to the pre-built convolutional neural network, and mark the nodes corresponding to the initial path on the composite network, wherein the composite network includes a first network, a second network and a third network; The change allocation module is used to change the initial path according to the actual choice of the traveler, generate a change notification, and correspondingly obtain the change node of the change path on the composite network, obtain all the paths traversed during the change process according to the association relationship between the change node and the traffic plan, and list the alternative change paths according to the association relationship between the change type and the corresponding node, summarize the alternative change paths, and obtain the corresponding change decision set; The publishing and uploading module is used to upload the change decision set to a database, and publish the change decision set to provide it to the travelers for selection; The database is used to store historical traffic road information, the actual traffic road information and the change decision set.

[0005] Furthermore, the establishing of a composite network according to the real-time traffic road information comprises: A first network is established with the intersection corresponding to the automobile road information as a node and the road at the intersection as an edge, and the representation form thereof is: G1 (V1, E1, W1); A second network is established with bus stops corresponding to the bus road information as nodes and lines between the bus stops as edges, and the representation form thereof is: G2 (V2, E2, W2); A third network is established with subway stations corresponding to the subway road information as nodes and lines between the subway stations as edges, and the representation form thereof is: G3 (V3, E3, W3); Then the representation form of the composite network is: ; In the formula, G represents a composite network, G k Represents the network at different layers, V k represents the set of nodes in the network, E k is the set of connected edges, W k Represents a weighted adjacency matrix; k takes the value of 1, 2 or 3, each layer corresponds to a subnetwork, and there are interactive connections between different layers.

[0006] Furthermore, the traffic plan includes travelers, travel goals, navigation principles and financial principles, and the travelers, travel goals, navigation principles and financial principles in the traffic plan are parsed based on the TF-IDF algorithm, and the route is planned according to the parsed travelers, travel goals, navigation principles and financial principles.

[0007] Furthermore, the change notification includes: a change type, a change content set, and a set of relationships between nodes in the change content set, which are marked as R1, R2, and R3 respectively; Among them, the change type R1 is to change the subnetwork type in the composite network; the change content set R2 is the content of the path change of the initial path; the set R3 of the relationship between the nodes in the change content set includes the relationship between nodes and the relationship between nodes and the whole.

[0008] Further, when the first node of the initial path changes, the second node with the change is tracked on the composite network, and the composite network is updated according to the first node and the second node; wherein the first node is the node that deviates from the initial path for the first time, and the second node is the node for re-path planning when deviating from the initial path; the first node and the second node are compared to obtain the influence range before and after the composite network is updated; the process of obtaining the influence range is: Build a rule engine to extract the navigation principles and financial principles from the complex network and define them as key factors. Then assign corresponding weights and scores to the key factors, and keep the weights and scores synchronized. Navigation principle: Car travel is assigned a weight of 3 > bus travel is assigned a weight of 2 > subway travel is assigned a weight of 1; Financial principle: The fare of the transportation mode is assigned a weight of 3 > the transfer fee of the transportation mode is assigned a weight of 2 > the time cost of the transportation mode is assigned a weight of 1; Multiply the score of each factor by its weight, and then add up all the results to generate the influence; compare the difference in influence before and after the update, perform interval comparison based on the influence difference, and obtain the corresponding influence range: where the influence difference = influence after update - influence before update.

[0009] Furthermore, the process of obtaining the association relationship between the change node and the traffic plan is as follows: Acquire a historical path corresponding to the current node according to the changed node, and create a first time label according to the historical path; Acquire a live path according to the city mathematical model map corresponding to the current node, and create a second time tag according to the live path; According to the first time tag and the second time tag, combined with the preset time constraints, the historical paths and live paths that meet the time constraints are classified to form several data groups, and association relationships are established within the several data groups with time as the connection point.

[0010] Further, the listing of candidate change paths according to the association relationship between the change type and the corresponding node includes: Through the change type corresponding to the change node, the association relationship between the change node and the traffic plan is obtained, the composite network is traversed, all candidate path sets are generated, and combined with the impact range generated during the change process, a rule engine is built to filter the path, and all generated alternative change paths are sorted and published.

[0011] Furthermore, in the convolutional neural network, the road impedance of the road section corresponding to each path is calculated according to the association relationship, and the formula of the road impedance is: ; Where minZ represents the minimum integral accumulation value of the impedance function of each path corresponding to the road section, η is any road section, and yx is the influence degree; The corresponding flow rate under any road section is preset as traffic Z , expressed as: ; In the formula, traffic Z Indicates the total flow of the road section, traffic1 indicates the flow of the car road section, traffic2 indicates the flow of the bus road section, and traffic3 indicates the flow of the subway road section; The formula for calculating the road impedance is used as the connection function of the preset convolutional neural network, and the number of nodes in the last hidden layer in the convolutional neural network model is consistent with the number of parameters involved in the road impedance formula. The generated path is used as the output value of the convolutional neural network, and the weight is obtained according to the road section flow solution formula. The weight of the last layer of the convolutional neural network is initialized, and the gradient descent algorithm is used for learning, and the entire network model is retrained.

[0012] A smart information release management method for urban traffic, comprising the following steps: S1, collecting real-time traffic road information of urban traffic according to a pre-configured urban mathematical model map, wherein the real-time traffic road information at least includes automobile road information, bus road information and subway road information; S2, establishing a composite network according to the real-time traffic road information, inputting the traveler's traffic plan into the composite network, generating an initial path based on the A* algorithm according to a pre-constructed convolutional neural network, and marking the nodes corresponding to the initial path on the composite network, wherein the composite network includes a first network, a second network, and a third network; S3, changing the initial path according to the actual choice of the traveler, generating a change notification, and correspondingly obtaining the change node of the change path on the composite network, obtaining all the paths traversed during the change process according to the association relationship between the change node and the traffic plan, and listing the alternative change paths according to the association relationship between the change type and the corresponding node, summarizing the alternative change paths, and obtaining a corresponding change decision set; S4, uploading the change decision set to a database, and publishing the change decision set to provide to the travelers for selection.

[0013] (III) Beneficial effects The present invention provides a smart information release management system and method for urban transportation, which has the following beneficial effects: 1. The present invention designs a composite network, including a first network, a second network and a third network. The three sub-networks are independent of the composite network, and are interconnected and interact with each other. According to the composite network, the transportation modes of car, bus and subway used in the travel process are combined, providing travelers with diversified transportation options and a richer combination of transportation modes, thereby improving the accuracy and practicality of the division of transportation modes; 2. The present invention integrates financial principles, navigation principles and time into the path analysis process by using a convolutional neural network, which not only improves the effectiveness of the operation process, but also facilitates the operation in the system and improves resource utilization. In this process, by using the formula for calculating road impedance as the connection function of the convolutional neural network, the convolutional neural network is improved to a certain extent, and the training speed of the convolutional neural network parameter migration is enhanced, so that the training speed and solution accuracy are good when solving complex traffic path problems. 3. The present invention obtains all paths and alternative paths of travelers in the process of changing the initial path through change notification, combined with the association between the change node and the traffic plan and the association between the change type and the corresponding node, and promptly replans the deviated path and provides a corresponding set of change decisions for travelers to choose, thereby improving the traceability of the system and enhancing the management ability of public services. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 A schematic diagram of a smart information release management system module for urban transportation according to an exemplary embodiment; Figure 2 A schematic flow chart of a method for managing smart information publishing for urban transportation is shown according to an exemplary embodiment. DETAILED DESCRIPTION

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

[0016] Embodiment 1: This embodiment provides a smart information release and management system for urban transportation. Figure 1 FIG. 1 is a schematic diagram of a smart information release management system for urban transportation according to an exemplary embodiment. Figure 1 The management system includes: a database, a dynamic collection module, a city map module, a demand generation module, a change allocation module and a publishing and uploading module, and the database, the dynamic collection module, the city map module, the demand generation module, the change allocation module and the publishing and uploading module are connected in communication; The following are specific descriptions and explanations of each module or component: City Map Module: Based on the OSM map model, and with the real city map as the background, the city mathematical model map is established by adding traffic hub intersections, path starting points, and path end points; Dynamic acquisition module: According to the urban mathematical model map, real-time traffic road information of urban traffic is collected; wherein the real-time traffic road information includes automobile roads, bus roads and subway roads; Demand Generation Module: A composite network is established according to the real-time traffic road information, including a first network, a second network and a third network; the process of establishing the composite network is: The first network is established with the intersections corresponding to the automobile roads as nodes and the roads at the intersections as edges, and its representation form is: G1 (V1, E1, W1); The second network is established with the bus stops corresponding to the bus roads as nodes and the lines between the bus stops as edges, and its representation form is: G2 (V2, E2, W2); The third network is established with the subway stations corresponding to the subway roads as nodes and the lines between the subway stations as edges. Its representation form is: G3 (V3, E3, W3); The representation of the composite network is: ; In the formula, G represents a composite network, G k Represents the network at different layers, V k represents the set of nodes in the network, E k is the set of connected edges, W k Represents a weighted adjacency matrix; k is 1, 2, or 3, and each layer corresponds to a subnetwork, and there are interactive connections between different layers; in, ; In the formula, A k is the set of edges between two nodes, represents the adjacency matrix, r represents the existence of a connection edge between two nodes, and R represents the absence of a connection edge between two nodes. represents the weighted adjacency matrix, i and j represent two nodes; By inputting the traffic plan of the traveler into the composite network, a convolutional neural network is pre-built, and an initial path is generated based on the A* algorithm, and the nodes corresponding to the initial path on the composite network are marked; wherein the traffic plan includes the traveler, travel goal, navigation principle and financial principle, and the traveler, travel goal, navigation principle and financial principle in the traffic plan are parsed based on the TF-IDF algorithm; The following are the explanations of the terms involved in the traffic plan: Travelers: represents different types of travelers; Travel destination: indicates the starting point, destination and purpose of the trip; Navigation principle: Indicates that during the trip, different modes of travel (such as car, bus and subway) are used to achieve the shortest route, avoid congestion during peak hours, etc. Financial principle: expresses the cost of using a mode of transport; Before TF-IDF calculation, the text data is segmented, stop words are removed, stemming or part-of-speech is restored; TF-IDF is used to calculate each word to evaluate their importance in different categories (travelers, travel goals, navigation principles and financial principles); through the TF-IDF calculation results, the most representative words or phrases are extracted to summarize travelers, travel goals, navigation principles and financial principles; Example: Travelers: Based on words with high TF-IDF values, this may include “commuters,” “tourists,” “elderly people,” “students,” etc. Travel purpose: may include "work", "shopping", "travel", "medical treatment", etc.; Navigation principles: may include “shortest path”, “fastest path”, “avoiding congestion”, “safety”, etc.; Financial principles: may include “minimum fare”, “budget constraints”, “public transport subsidies”, etc.; Plan routes based on analyzed travelers, travel goals, navigation principles, and financial principles; In this embodiment, the A* algorithm is also introduced for path planning. The A* algorithm is a widely used graph search algorithm that combines the advantages of the best first search and the Dijkstra algorithm. It can effectively find the shortest path from the travel start point to the travel end point. The general cost estimation function f(n) is defined according to the travel start point start and the travel end point end, which is expressed as: ; In the formula, n represents the current location node, g(n) represents the actual moving distance from the starting point to the current location, and h(n) represents the estimated moving cost from the current point to the end point of the trip, also known as the heuristic function, which is usually calculated using Manhattan distance or Euclidean distance; Change the allocation module: The initial path is changed according to the actual choice of the traveler, and a change notification is generated. The change node in the composite network is obtained accordingly. According to the association between the change node and the traffic plan, all the paths traversed during the change process are obtained, and the alternative change paths are listed according to the association between the change type and the corresponding node. The alternative change paths are summarized to obtain a corresponding change decision set; Receive a change notification, which includes a change type, a change content set, and a set of relationships between nodes in the change content set, which are marked as R1, R2, and R3 respectively; wherein the change type R1 is a subnetwork type in a changed composite network; the change content set R2 is the content of the path change to the initial path; the set R3 of relationships between nodes in the change content set includes relationships between nodes and relationships between nodes and the whole; R1 specifically includes: changing the first network to the second network, changing the second network to the first network, changing the first network to the third network, changing the third network to the first network, changing the second network to the third network, and changing the third network to the second network; R2 specifically includes: deleting the front node and the back node of the initial path on the composite network, deleting the relationship between the two, and adding nodes on the composite network, adding attributes to the front node, reducing attributes of the front node, changing attributes of the front node, adding attributes to the back node, reducing attributes of the back node, and changing attributes of the back node; it should be noted that the attributes correspond to the travel goals, navigation principles, and financial principles in the traffic plan; R3 specifically includes: adjacent relations, path relations, connection relations, transfer relations, and influence relations; A detailed explanation of the relationship involved in R3: Adjacent relationship: In any sub-network, there is a direct passage between two nodes; for example, in the first network, there is a direct passage relationship between two adjacent intersection nodes; Path relationship: represents the relationship between two nodes in any sub-network; for example, the route between two bus stops in the second network; Connection relationship: represents the line connection between different nodes in any sub-network; for example, in the third network, different subway stations are connected by subway lines; Transfer relationship: Indicates that in a composite network, a node serves as a transfer point between different sub-networks; Influence relationship: In a composite network, the state of a node may affect the nodes of other sub-networks; When the first node of the initial path changes, the second node with the change is tracked on the composite network, and the composite network is updated according to the first node and the second node; wherein the first node is the node that deviates from the initial path for the first time, and the second node is the node for re-path planning when deviating from the initial path; The analysis process of re-planning the path after deviating from the initial path is as follows: First, obtain the initial path, which is numbered as j0 and put into the path candidate set J, at which point i=1; Secondly, get the last path j in the path candidate set J i As the current path, the path j i All nodes d except the end point k As potential deviation nodes, use the A* algorithm to calculate the k The shortest distance to the end point is taken as the deviation from the shortest path. In order to avoid the sum of the previous paths, it is necessary to hide (dk , end) part of the path; Then, the deviated shortest path is compared with the non-deviated path (start, d k ) as one of the candidate paths for the i-th shortest path, and put the path into the path candidate set J; Finally, the path with the shortest length is selected from the path candidate set J as the i+1th shortest path j i+1 , and put the path into the path candidate set J; Compare the first node and the second node according to the path candidate set J to obtain the influence range before and after the composite network is updated; The process of obtaining the impact range is: Build a rule engine to extract the navigation principles and financial principles from the complex network, define them as key factors, and assign corresponding weights and scores to the key factors, and keep the weights and scores synchronized; Navigation principle: Car travel is assigned a weight of 3 > bus travel is assigned a weight of 2 > subway travel is assigned a weight of 1; Financial principle: The fare of the transportation mode is assigned a weight of 3 > the transfer fee of the transportation mode is assigned a weight of 2 > the time cost of the transportation mode is assigned a weight of 1; Multiply the score of each factor by its weight and add up all the results to generate the impact. Compare the difference in impact before and after the update to obtain the corresponding impact range: where impact difference = impact after update − impact before update. It should be noted that the impact range usually refers to the magnitude of the impact change and its impact on the overall system or model. To a certain extent, it affects the efficiency, travel time, congestion and overall traffic flow of path planning. It is usually analyzed and explained using percentage changes, which are expressed as the ratio of the impact difference to the impact before the update. Define the preset interval based on the percentage change. For example, if the preset interval is [5%, 10%], then: When the impact is less than 5%, the impact at this time is marked as 1 and assigned a first-level character. 1 and the first-level character are combined to generate the first impact range, indicating that the impact before and after the update is small; When the impact is between [5%, 10%], the impact is marked as 2 and assigned a secondary character. 2 is combined with the primary character to generate the second impact range, indicating that the impact before and after the update is average. When the impact degree is greater than 10%, the impact degree is marked as 3 and assigned a third-level character. 3 and the third-level character are combined to generate the third impact range, indicating that the impact before and after the update is greater. This analysis process can help understand the changes brought about by the update node, thereby providing a basis for subsequent decision-making and optimization. The process of obtaining the association between the change node and the traffic plan is as follows: Acquire a historical path corresponding to the current node according to the changed node, create a first time tag according to the historical path, acquire a live path according to the city mathematical model map corresponding to the current node, create a second time tag according to the live path, classify the historical paths and live paths that meet the time constraints according to the first time tag and the second time tag, combined with preset time constraints, to form a number of data groups, and establish association relationships within the several data groups with time as a connection point; By matching the change type to the change node, we can obtain the relationship between the change node and the traffic plan, traverse the composite network, generate a set of all candidate paths, and build a rule engine to filter the paths based on the impact range generated during the change process. We can then organize and publish all the generated candidate change paths. In the convolutional neural network, the road impedance of each path corresponding to the road section is calculated according to the association relationship. In this embodiment, the road impedance is a measure of the total resistance that the corresponding vehicle in the composite network needs to overcome when traveling on a certain road section. It is usually expressed in terms of time, flow, distance or cost. Here, the road section flow parameter is selected for analysis, and the formula based on the road impedance is: (1) In formula (1), minZ represents the minimum value of the integral accumulation of the impedance function of each path corresponding to the road section, η is any road section, and yx is the influence degree; The corresponding flow rate under any road section is preset as traffic Z , expressed as: (2) In formula (2), traffic Z Indicates the total flow of the road section, traffic1 indicates the flow of the car road section, traffic2 indicates the flow of the bus road section, and traffic3 indicates the flow of the subway road section; The formula for calculating the road impedance is used as a connection function of a preset convolutional neural network, and the number of nodes in the last hidden layer of the convolutional neural network model is consistent with the number of parameters involved in the road impedance formula, and the generated path is used as the output value of the convolutional neural network, and the weight is obtained according to the road section flow solution formula, the weight of the last layer of the convolutional neural network is initialized, and the gradient descent algorithm is used for learning, and the entire network model is retrained; For the first network, its mathematical model is as follows: ; In the formula, represents the weighted neighbor matrix of the first network, represents the adjacency matrix of the first network, Indicates the length of the road section between the nodes. Indicates the speed limit of the road section between intersections. Indicates the real-time traffic flow of the road section. Indicates the capacity of the road section; m and n are both conversion coefficients of the formula, and m is usually taken as 0.15 and n is usually taken as 4; For the second network, its mathematical model is as follows: ; In the formula, represents the weighted neighbor matrix of the second network, represents the adjacency matrix of the second network, represents the traffic flow on the road section between bus stops, Indicates the departure frequency between two bus stops; For the third network, its mathematical model is as follows: ; In the formula, represents the weighted neighbor matrix of the third network, represents the adjacency matrix of the third network, represents the length of the road section between two subway stations. Indicates the running speed between two subway stations. represents the traffic flow of the section between two subway stations. Indicates the departure frequency between two subway stations; Release upload module: Upload the change decision set to the database and publish the change decision set for travelers to choose database: The historical traffic road information, the live traffic road information and the change decision set are stored.

[0017] In summary, the present invention includes a database, a dynamic collection module, a city map module, a demand generation module, a change allocation module and a publishing and uploading module. By collecting real-time traffic road information of urban traffic, a composite network is constructed, including a first network, a second network and a third network. By analyzing the traffic plan of travelers, considering multiple factors such as time, navigation principles and financial principles, and combining the A* algorithm to generate a corresponding initial path, when the traveler deviates from the initial path, a change notification is generated, and the path planning is re-performed; at the same time, the impact range of each change and the corresponding road section flow on each sub-network are recorded, the road impedance is calculated as a connection function, and the generated path is used as an output value. By timely updating the composite network, a rich variety of traffic options are continuously provided for travelers to choose; in this process, by monitoring the association relationship generated between nodes, corresponding analysis is performed through the association relationship, and to a certain extent, the management efficiency is improved.

[0018] Embodiment 2: This embodiment provides a smart information publishing and management method for urban transportation. Figure 2 is a schematic diagram of a module of a smart information publishing management method for urban traffic according to an exemplary embodiment. Figure 2 , including the following steps: S1, collecting real-time traffic road information of urban traffic according to a pre-configured urban mathematical model map, wherein the real-time traffic road information at least includes automobile road information, bus road information and subway road information; S2, establishing a composite network according to the real-time traffic road information, inputting the traveler's traffic plan into the composite network, generating an initial path based on the A* algorithm according to a pre-constructed convolutional neural network, and marking the nodes corresponding to the initial path on the composite network, wherein the composite network includes a first network, a second network, and a third network; S3, changing the initial path according to the actual choice of the traveler, generating a change notification, and correspondingly obtaining the change node of the change path on the composite network, obtaining all the paths traversed during the change process according to the association relationship between the change node and the traffic plan, and listing the alternative change paths according to the association relationship between the change type and the corresponding node, summarizing the alternative change paths, and obtaining a corresponding change decision set; S4, uploading the change decision set to a database, and publishing the change decision set to provide to the travelers for selection.

[0019] The size of the interval and threshold is set to facilitate comparison. The size of the threshold depends on the amount of sample data and the number of bases set by technical personnel in this field for each group of sample data; as long as it does not affect the proportional relationship between the parameter and the quantized value.

[0020] In the application, the several formulas involved are all calculated by removing dimensions and taking their numerical values, and the formula is a formula obtained by collecting a large amount of data and performing software simulation to obtain the most recent real situation. The formula is set by technical personnel in this field according to actual conditions.

[0021] The above embodiments may be implemented in whole or in part by software, hardware, firmware or any other combination thereof. When implemented using software, the above embodiments may be implemented in whole or in part in the form of a computer program product. A person of ordinary skill in the art may appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein may be implemented in electronic hardware, or in a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution.

[0022] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, and may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0023] The above description is only a specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application.

Claims

1. A smart information release and management system for urban transportation, characterized in that: The system includes: a database, a dynamic collection module, a city map module, a demand generation module, a change allocation module, and a publishing and uploading module; The city map module is used to establish a city mathematical model map based on the OSM map model and the real city map as the background, adding traffic hub intersections, path starting points, and path end points; The dynamic acquisition module is used to collect real-time traffic road information of urban traffic according to the urban mathematical model map, wherein the real-time traffic road information at least includes automobile road information, bus road information and subway road information; The demand generation module is used to establish a composite network according to the real-time traffic road information, input the traveler's traffic plan into the composite network, generate an initial path based on the A* algorithm according to the pre-built convolutional neural network, and mark the nodes corresponding to the initial path on the composite network, wherein the composite network includes a first network, a second network and a third network; The change allocation module is used to change the initial path according to the actual choice of the traveler, generate a change notification, and correspondingly obtain the change node of the change path on the composite network, obtain all the paths traversed during the change process according to the association relationship between the change node and the traffic plan, and list the alternative change paths according to the association relationship between the change type and the corresponding node, summarize the alternative change paths, and obtain the corresponding change decision set; The publishing and uploading module is used to upload the change decision set to a database, and publish the change decision set to provide it to the travelers for selection; The database is used to store historical traffic road information, the actual traffic road information and the change decision set.

2. According to claim 1, a smart information release and management system for urban transportation is characterized by: The step of establishing a composite network according to the real-time traffic road information comprises: A first network is established with the intersection corresponding to the automobile road information as a node and the road at the intersection as an edge, and the representation form thereof is: G1 (V1, E1, W1); A second network is established with bus stops corresponding to the bus road information as nodes and lines between the bus stops as edges, and the representation form thereof is: G2 (V2, E2, W2); A third network is established with subway stations corresponding to the subway road information as nodes and lines between the subway stations as edges, and the representation form thereof is: G3 (V3, E3, W3); Then the representation form of the composite network is: ; In the formula, G represents a composite network, G k Represents the network at different layers, V k represents the set of nodes in the network, E k is the set of connected edges, W k Represents a weighted adjacency matrix; k takes the value of 1, 2 or 3, each layer corresponds to a subnetwork, and there are interactive connections between different layers.

3. According to claim 1, a smart information release and management system for urban transportation is characterized in that: The transportation plan includes travelers, travel goals, navigation principles and financial principles, and the travelers, travel goals, navigation principles and financial principles in the transportation plan are parsed based on the TF-IDF algorithm, and the route is planned according to the parsed travelers, travel goals, navigation principles and financial principles.

4. According to claim 3, a smart information release and management system for urban transportation is characterized in that: The change notification includes: a change type, a change content set, and a set of relationships between nodes in the change content set, which are marked as R1, R2, and R3 respectively; Among them, the change type R1 is to change the subnetwork type in the composite network; the change content set R2 is the content of the path change of the initial path; the set R3 of the relationship between the nodes in the change content set includes the relationship between nodes and the relationship between nodes and the whole.

5. According to claim 4, a smart information release and management system for urban transportation is characterized in that: When the first node of the initial path changes, the second node with the change is tracked on the composite network, and the composite network is updated according to the first node and the second node; wherein the first node is a node that deviates from the initial path for the first time, and the second node is a node for re-path planning when deviating from the initial path; the first node and the second node are compared to obtain the influence range before and after the composite network is updated; the process of obtaining the influence range is: Build a rule engine to extract the navigation principles and financial principles from the complex network and define them as key factors. Then assign corresponding weights and scores to the key factors, and keep the weights and scores synchronized. Navigation principle: Car travel is assigned a weight of 3 > bus travel is assigned a weight of 2 > subway travel is assigned a weight of 1; Financial principle: The fare of the transportation mode is assigned a weight of 3 > the transfer fee of the transportation mode is assigned a weight of 2 > the time cost of the transportation mode is assigned a weight of 1; Multiply the score of each factor by its weight, and then add up all the results to generate the influence; compare the difference in influence before and after the update, perform interval comparison based on the influence difference, and obtain the corresponding influence range: where the influence difference = influence after update - influence before update.

6. According to claim 1, a smart information release and management system for urban transportation is characterized by: The process of obtaining the association relationship between the change node and the traffic plan is as follows: Acquire a historical path corresponding to the current node according to the changed node, and create a first time label according to the historical path; Acquire a live path according to the city mathematical model map corresponding to the current node, and create a second time tag according to the live path; According to the first time tag and the second time tag, combined with the preset time constraints, the historical paths and live paths that meet the time constraints are classified to form several data groups, and association relationships are established within the several data groups with time as the connection point.

7. According to claim 6, a smart information release and management system for urban transportation is characterized in that: The listing of candidate change paths according to the association relationship between the change type and the corresponding node includes: Through the change type corresponding to the change node, the association relationship between the change node and the traffic plan is obtained, the composite network is traversed, all candidate path sets are generated, and combined with the impact range generated during the change process, a rule engine is built to filter the path, and all generated alternative change paths are sorted and published.

8. A smart information release and management system for urban transportation according to claim 6 or 7, characterized in that: In the convolutional neural network, the road impedance of the road section corresponding to each path is calculated according to the association relationship, and the formula of the road impedance is: ； In the formula, minZ represents the minimum integral accumulation value of the impedance function of each path corresponding to the road section, η is any road section, and yx is the influence degree; The corresponding flow rate under any road section is preset as traffic Z , expressed as: ； In the formula, traffic Z Indicates the total flow of the road section, traffic1 indicates the flow of the car road section, traffic2 indicates the flow of the bus road section, and traffic3 indicates the flow of the subway road section; The formula for calculating the road impedance is used as the connection function of the preset convolutional neural network, and the number of nodes in the last hidden layer in the convolutional neural network model is consistent with the number of parameters involved in the road impedance formula. The generated path is used as the output value of the convolutional neural network, and the weight is obtained according to the road section flow solution formula. The weight of the last layer of the convolutional neural network is initialized, and the gradient descent algorithm is used for learning, and the entire network model is retrained.

9. A smart information release management method for urban traffic, using any system described in claims 1 to 8, characterized in that: The steps include: S1, collecting real-time traffic road information of urban traffic according to a pre-configured urban mathematical model map, wherein the real-time traffic road information at least includes automobile road information, bus road information and subway road information; S2, establishing a composite network according to the real-time traffic road information, inputting the traveler's traffic plan into the composite network, generating an initial path based on the A* algorithm according to a pre-constructed convolutional neural network, and marking the nodes corresponding to the initial path on the composite network, wherein the composite network includes a first network, a second network, and a third network; S3, changing the initial path according to the actual choice of the traveler, generating a change notification, and correspondingly obtaining the change node of the change path on the composite network, obtaining all the paths traversed during the change process according to the association relationship between the change node and the traffic plan, and listing the alternative change paths according to the association relationship between the change type and the corresponding node, summarizing the alternative change paths, and obtaining a corresponding change decision set; S4, uploading the change decision set to a database, and publishing the change decision set to provide to the travelers for selection.

Images