AI-based road network planning methods and systems

Abstract

This application relates to the field of artificial intelligence technology, providing a road network planning method and system based on AI decision-making. The method includes: extracting multi-source datasets from traffic monitoring centers using data crawling technology and merging them with urban road network maps to obtain a road network topology map; constructing a data processing terminal by performing point matching, including multiple partitions for processing traffic key data; setting a road network optimization space based on data correlation and traffic quality assessment indicators, clustering and constructing a road network planning model; integrating the road network optimization space and planning model into the data processing terminal to obtain multiple road network planning schemes; and determining the first road network planning scheme based on multi-scheme simulation and evaluation. This application solves the technical problem that traditional road network planning methods struggle to make efficient and accurate decisions when facing complex and ever-changing traffic environments and demands, achieving intelligent road network planning based on AI, and improving the operational efficiency and safety of road networks.

Description

Technical Field

[0001] This application relates to the field of artificial intelligence technology, specifically to road network planning methods and systems based on AI decision-making. Background Technology

[0002] With the accelerating pace of urbanization, road transportation, as a crucial support for urban development, faces unprecedented challenges in its construction and management. The rapid growth of urban populations and the continuous increase in vehicle ownership have led to increasingly prominent problems such as traffic congestion and frequent traffic accidents, causing considerable inconvenience to travel. Summary of the Invention

[0003] This application provides a road network planning method and system based on AI decision-making, aiming to solve the technical problem that traditional road network planning methods are unable to make efficient and accurate decisions when faced with complex and ever-changing traffic environments and demands.

[0004] In view of the above problems, this application provides a road network planning method and system based on AI decision-making.

[0005] The first aspect disclosed in this application provides a road network planning method based on AI decision-making. The method includes: extracting multi-source datasets from a traffic monitoring center using data scraping technology, and fusing the multi-source datasets with an urban road network map to obtain a road network topology map, the road network topology map containing multiple traffic points; performing point matching based on the road network topology map, constructing a data processing terminal based on the point matching results, the data processing terminal containing multiple data processing partitions, the multiple data processing partitions being responsible for data processing corresponding to the traffic points; setting a road network optimization space, the road network optimization space being divided into data clusters based on data correlation and traffic quality assessment indicators to obtain clustering analysis results; constructing a road network planning model based on the clustering analysis results and the road network topology map; integrating the road network optimization space and the road network planning model into the multiple data processing partitions of the data processing terminal to perform road network planning and obtain multiple planning schemes; and performing simulation evaluation based on the multiple planning schemes to obtain a first road network planning scheme.

[0006] Another aspect of this application discloses an AI-based road network planning system, comprising: a data fusion module, which extracts multi-source datasets from a traffic monitoring center using data crawling technology, and performs multi-source data fusion with an urban road network map to obtain a road network topology map, the road network topology map containing multiple traffic points; a point matching module, which performs point matching based on the road network topology map, and constructs a data processing terminal based on the point matching results, the data processing terminal containing multiple data processing partitions, the multiple data processing partitions being responsible for data processing corresponding to the traffic points; and clustering. The system comprises the following modules: a partitioning module, which sets up a road network optimization space, and performs data clustering based on data correlation and traffic quality assessment indicators to obtain clustering analysis results; a model building module, which constructs a road network planning model based on the clustering analysis results and the road network topology diagram; a road network planning module, which integrates the road network optimization space and the road network planning model into multiple data processing partitions of the data processing terminal to perform road network planning and obtain multiple planning schemes; and a simulation evaluation module, which performs simulation evaluation based on the multiple planning schemes to obtain a first road network planning scheme.

[0007] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0008] The aforementioned AI-based road network planning method utilizes data scraping technology to acquire multi-source datasets from traffic monitoring centers. These datasets, along with urban road network maps, are fused to obtain a road network topology map. Subsequently, point matching is performed based on this topology map, and a dedicated data processing partition is constructed for each traffic node to process data related to those nodes. Next, a road network optimization space is established, and cluster analysis is performed on this space using data correlation and traffic quality assessment indicators to better understand traffic conditions. Then, based on the cluster analysis results and the road network topology map, a road network planning model is constructed. This model and the road network optimization space are integrated into the previously established data processing partitions for road network planning, resulting in multiple planning schemes. Finally, these schemes are simulated and evaluated to select the optimal road network planning scheme. This process fully leverages data scraping and fusion technologies, combined with intelligent road network optimization and planning models, effectively improving the efficiency and accuracy of road network planning.

[0009] The above description is merely an overview of the technical solution of this application. In order to better understand the technical means of this application and to implement it in accordance with the contents of the specification, and to make the above and other objects, features and advantages of this application more obvious and understandable, specific embodiments of this application are given below. Attached Figure Description

[0010] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0011] Figure 1 This is a flowchart illustrating a road network planning method based on AI decision-making in one embodiment;

[0012] Figure 2 This is an architecture diagram of a road network planning system based on AI decision-making in one embodiment.

[0013] Figure labeling: Data fusion module 1, point matching module 2, clustering module 3, model building module 4, road network planning module 5, simulation evaluation module 6. Detailed Implementation

[0014] This application provides a road network planning method and system based on AI decision-making, which solves the technical problem that traditional road network planning methods are unable to make efficient and accurate decisions when faced with complex and ever-changing traffic environments and demands.

[0015] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0016] It should be noted that the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units that are explicitly listed, but may include other steps or modules that are not explicitly listed or that are inherent to such process, method, product, or device.

[0017] Example 1

[0018] like Figure 1 As shown, this application provides a road network planning method based on AI decision-making, the method comprising:

[0019] Based on data crawling technology, multi-source datasets are extracted from traffic monitoring centers, and multi-source data fusion is performed based on the multi-source datasets and urban road network maps to obtain a road network topology map, which contains multiple traffic points.

[0020] With the rapid advancement of artificial intelligence technology and the increasing abundance of traffic data resources, the application of AI in road network planning will continue to deepen, demonstrating enormous development potential and value. AI-driven road network planning can achieve precise analysis and processing of traffic data, predict traffic flow, identify traffic bottlenecks, and adjust traffic organization strategies in real time through intelligent algorithms, thereby improving the intelligence level of road network planning.

[0021] In this embodiment, the system terminal successfully acquired datasets from multiple sources from the traffic monitoring center using data crawling technology. These datasets were combined with an urban road network map, and after multi-source data fusion processing, the fusion results were graded and weighted to obtain a detailed road network topology map. This road network topology map not only displays the overall layout of urban roads but also marks several key traffic points. These traffic points are crucial for subsequent traffic planning and management because they are typically important road sections or nodes with high traffic volume, prone to congestion or accidents. This topology map provides a more intuitive understanding of the city's traffic operation, offering strong support for subsequent traffic optimization.

[0022] Furthermore, this application provides a method for extracting multi-source datasets from traffic monitoring centers based on data crawling technology, and fusing multi-source data based on urban road network maps to obtain a road network topology map. The method also includes:

[0023] Read the city road network map to obtain the multiple traffic points;

[0024] The multi-source dataset is split to obtain multiple split datasets, and the multiple split datasets are time-series converted to obtain multiple key point datasets;

[0025] Based on the multiple traffic points, the datasets of the multiple traffic points are matched, and the data in the same region are fused according to the data matching results to obtain the fusion results of multiple regions.

[0026] Preferably, the system terminal reads the city road network map, accurately identifying and extracting multiple key traffic points. These traffic points are crucial nodes in the urban transportation system, essential for traffic flow control and optimization. Subsequently, the collected multi-source datasets are split, refining them into multiple sub-datasets. This facilitates a deeper analysis of the characteristics and correlations of each dataset. Next, timestamp fields are identified and extracted from these sub-datasets. Based on the extracted timestamps, the data items in the sub-datasets are sorted to ensure they are arranged in chronological order. Then, the time-series-processed datasets are stored, forming multiple key point datasets. Time-series processing allows the system terminal to understand data trends over time, providing crucial information for subsequent decision-making. Finally, data matching is performed on the key point datasets using the previously identified traffic points. Through matching, datasets belonging to the same region are found and fused. This step helps integrate data from different sources within the same region, obtaining more comprehensive and accurate regional fusion results. These regional fusion results provide important data support for subsequent traffic planning and management.

[0027] Furthermore, this application provides a method for extracting multi-source datasets from traffic monitoring centers based on data crawling technology, and fusing multi-source data based on urban road network maps to obtain a road network topology map. The method also includes:

[0028] Based on the fusion results of the multiple regions, regional accident statistical analysis is performed, and a level assessment is conducted in conjunction with the regional classification standards. Based on the assessment results, multi-level nodes are determined.

[0029] Based on the regional fusion results, road attribute information is extracted, and multi-level node connections are performed based on the road attribute information to construct an initial road network topology diagram.

[0030] Optionally, based on the fusion results of multiple regions, the system terminal first conducts in-depth statistical analysis of regional accidents. This step aims to statistically determine the frequency and severity of accidents in each region through data and facts, enabling the system terminal to comprehensively understand the accident situation in each region, including key indicators such as accident frequency and severity. Subsequently, in conjunction with regional classification standards, each region is graded, i.e., based on accident frequency and severity, all regions are divided into Level 1, Level 2, and Level 3 accident-prone areas. This step helps to more clearly understand the traffic conditions and accident risks in different regions, providing an important basis for subsequent decision-making. Next, based on the regional grading results, the locations of multi-level nodes are determined. These nodes play an important role in the transportation network, serving as key points for connecting different regions and managing traffic flow, or as accident-prone areas. Determining multi-level nodes lays the foundation for building a more rational and efficient transportation network. Then, the system terminal extracts road attribute information from the regional fusion results, including key parameters such as road length, width, and capacity. Based on this road attribute information, the multi-level nodes are connected, thereby constructing an initial road network topology map. This diagram visually illustrates the layout of the road network and the connections between nodes, providing strong support for subsequent optimization and planning. In summary, through the analysis and processing of the regional fusion results, successful accident statistical analysis, regional level assessment, multi-level node determination, and the construction of a road network topology map were achieved. This work not only deepens the understanding of urban traffic conditions but also provides important data support and decision-making basis for subsequent traffic optimization and decision-making.

[0031] The distribution weights of the regional fusion results are calculated to obtain the distribution weight calculation results, which include traffic flow weights, congestion index weights, and accident frequency weights.

[0032] Based on the distribution weight calculation results, the weights are integrated, and corresponding weights are assigned to the initial road network topology diagram according to the integrated results to obtain the road network topology diagram.

[0033] Optionally, the system terminal performs distributed weight calculations on the regional fusion results. This step aims to obtain the weights of each region on key indicators such as traffic flow, congestion index, and accident frequency. These weights reflect the importance and influence of each region within the transportation network, providing crucial information for subsequent decision-making. Specifically, the system terminal first extracts records of traffic flow, congestion index, and accident frequency for each region over a past period from the regional fusion results. Then, it sums the traffic flow and accident frequency data for all regions to obtain the total traffic flow and total accident frequency, respectively. Simultaneously, it sums the congestion index data and divides it by the number of regions to obtain the average congestion index. Subsequently, for each region, it calculates the proportion of traffic flow to total traffic flow to obtain the region's traffic flow weight. Then, it calculates the ratio of the region's congestion index to the average congestion index to obtain the region's congestion index weight. Finally, it calculates the proportion of each region's accident frequency to the total accident frequency to obtain the region's accident frequency weight. After obtaining the distributed weight results, the system terminal performs weight integration processing. This step involves comprehensively considering the weights of different indicators to obtain a comprehensive weight value, which is achieved by summing and averaging the traffic flow weight, congestion index weight, and accident frequency weight for each region. This comprehensive weight value can more comprehensively reflect the overall situation of each region in the traffic network. Next, the initial road network topology map is weighted according to the comprehensive weight result. This means that the system terminal assigns different weights to corresponding nodes and road segments in the road network topology map based on the actual influence of each region in the traffic network. This helps to more accurately identify key nodes and road segments in the traffic network, providing strong support for subsequent optimization and planning. Finally, a weighted road network topology map is obtained. This map not only shows the layout and connection relationships of the road network but also reflects the importance and influence of each region and road segment in the traffic network through weight values. This will provide important reference for subsequent traffic decisions and planning.

[0034] Based on the road network topology map, point matching is performed, and a data processing terminal is constructed based on the point matching results. The data processing terminal contains multiple data processing partitions, and the multiple data processing partitions are responsible for the data processing of the corresponding traffic points.

[0035] In one embodiment, the system terminal first performs point matching based on the road network topology map. This step is mainly to accurately identify key points in the traffic network, such as intersections, tidal flow roads, congestion hotspots, and accident-prone areas. These points are of particular importance in traffic management and therefore require special attention. Subsequently, a data processing terminal is constructed based on the point matching results. This data processing terminal consists of multiple data processing partitions, each corresponding to a specific traffic point. This design allows the system terminal to allocate appropriate data processing resources based on the importance and complexity of different points. For traffic points, data processing partitions with higher computing power are allocated to ensure that the data from these key points can be processed quickly and accurately. For other ordinary points, although data processing is also performed, data processing partitions with relatively lower computing power are allocated to balance overall processing efficiency and resource utilization. In this way, an efficient and accurate data processing system is constructed, which can better support traffic management and decision-making. By prioritizing the processing of data from key points, traffic problems can be identified more quickly, corresponding solutions can be formulated, and the overall operational efficiency of the traffic network can be optimized.

[0036] A road network optimization space is set up, and the road network optimization space is divided into data clusters based on data correlation and traffic quality assessment indicators to obtain cluster analysis results;

[0037] In one embodiment, setting up a road network optimization space is a complex process, based on data correlation and traffic quality assessment indicators for data clustering. In this process, the system terminal first uses AI algorithms to perform correlation analysis on the fusion results of multiple regions, identifying the inherent connections and similarities between different regions. Then, based on these correlations and assessment indicators, the road network data is clustered, grouping regions with similar traffic characteristics or traffic problems into one category. Through this process, the results of the clustering analysis can be obtained, namely, the road network conditions and characteristics of different regions. Subsequently, the system terminal constructs multiple road network optimization spaces based on the segmented results, each responsible for a specific traffic characteristic or traffic problem. Such clustering analysis results provide strong data support for subsequent road network optimization and traffic management, helping to formulate more accurate traffic strategies and improve the overall operational efficiency of the road network.

[0038] Furthermore, this application provides a method for setting up a road network optimization space, wherein the road network optimization space is divided into data clusters based on data correlation and traffic quality assessment indicators, and the method further includes:

[0039] Based on the fusion results of the multiple regions, a regional traffic quality assessment is conducted to obtain the regional traffic quality assessment results.

[0040] Using AI algorithms, correlation analysis is performed on the fusion results of the multiple regions to calculate the correlation index between different regions and generate regional correlation.

[0041] Based on the regional traffic quality assessment results, the regional correlation, and the traffic quality assessment indicators, the multiple regional fusion results are clustered and divided to form multiple road network optimization spaces.

[0042] Optionally, based on the fusion results from multiple regions, the system terminal first conducts a comprehensive assessment of the traffic quality of each region, obtaining specific assessment results. These results reflect the actual performance of each region in terms of traffic flow, congestion, and accident rate. Subsequently, regional features are extracted from the fusion results of multiple regions, and AI algorithms are used to perform correlation analysis on the extracted results. This step aims to identify the inherent connections and mutual influences between different regions. By calculating correlation indicators, regional correlation degrees are generated, and these data reveal the similarities and differences in traffic conditions among regions. Finally, combining the regional traffic quality assessment results, regional correlation degrees, and traffic quality assessment indicators, the fusion results of multiple regions are clustered. This process groups regions with similar traffic characteristics and problems into one category, forming multiple road network optimization spaces. These spaces provide clear goals and directions for subsequent road network optimization and traffic management. Through this series of analyses and classifications, the system terminal can more accurately understand the traffic conditions of each region, identify existing problems and bottlenecks, and provide strong data support for formulating effective traffic strategies and measures.

[0043] Furthermore, this application provides a method for performing correlation analysis on the fusion results of the multiple regions, calculating correlation indicators between different regions, and generating regional correlation. The method also includes:

[0044] Regional features are extracted from the fusion results of the multiple regions to obtain a first regional feature set, which includes traffic flow features, road structure features, and population distribution features.

[0045] The first region feature set is relevance index is calculated using AI algorithms, and the calculation results are standardized to obtain the first relevance index set.

[0046] The first set of relevant indicators is traversed, and corresponding weights are assigned. A weighted average is then performed to generate the regional correlation degree.

[0047] Optionally, the system terminal performs regional feature extraction on the fusion results of multiple regions. This process involves multiple aspects such as traffic flow characteristics, road structure characteristics, and population distribution characteristics, thus forming a first regional feature set. This set provides the system terminal with a comprehensive and in-depth perspective on the characteristics of each region. Subsequently, the first regional feature set is input into the AI ​​algorithm. The system terminal uses the correlation analysis module in the AI ​​algorithm to calculate the correlation index of the input first regional feature set. The AI ​​algorithm has built-in Pearson correlation coefficient and Spearman rank correlation coefficient, and selects an appropriate correlation coefficient for correlation analysis based on the specific characteristics and needs of the data. During the calculation process, the AI ​​algorithm analyzes the linear relationship, monotonic relationship, etc., between various features and outputs the corresponding correlation coefficients. Afterwards, the calculated correlation indexes are standardized to eliminate the differences in dimensions and units between different indicators. Standardization can be performed using methods such as Z-score standardization and Min-Max standardization to convert the indicator values ​​into a standard normal distribution with a mean of 0 and a standard deviation of 1. Through standardization, it can be ensured that all indicators in the first correlation index set have the same scale, which facilitates subsequent data analysis and comparison. Then, the standardized correlation indicators were used as the first correlation indicator set, which contained correlation information between regional characteristics. This first correlation indicator set provided crucial data support for subsequent regional correlation analysis and road network optimization spatial division. Furthermore, the system terminal traversed the first correlation indicator set and assigned corresponding weights to each indicator. This step was based on an understanding of the importance of each indicator in regional correlation. Regional correlation data was successfully generated using a weighted average method. This data provides strong support for subsequent road network optimization and traffic management, helping the system terminal better understand and address traffic correlation issues between different regions.

[0048] Furthermore, this application provides a method that, before clustering the multiple regional fusion results based on the regional traffic quality assessment results, the regional correlation, and the traffic quality assessment indicators, further includes:

[0049] Based on the regional traffic quality assessment results, key traffic quality indicators are extracted to form a set of key traffic quality indicators.

[0050] Based on the aforementioned regional correlation, the interactions and influences between different regions are analyzed to form a set of regional correlation indicators.

[0051] Optionally, the system terminal, based on a comprehensive assessment of regional traffic quality, selects the most representative and critical traffic quality indicators, thus forming a key traffic quality indicator set. These indicators accurately reflect the traffic conditions of each region, providing a clear direction for subsequent optimization and management. Simultaneously, based on the calculated regional correlation degree, the interactions and influences between different regions are analyzed in depth. By constructing a regional correlation degree indicator set, the position and role of each region in the traffic network, as well as the degree of correlation and mutual influence patterns among them, can be clearly identified. In summary, by extracting key traffic quality indicators and forming a regional correlation degree indicator set, a deep understanding and comprehensive grasp of the regional traffic conditions are achieved. These indicator sets not only provide strong data support for traffic management but also offer important reference for formulating effective traffic optimization strategies.

[0052] By combining the aforementioned set of key traffic quality indicators, the traffic quality assessment indicators are further expanded to form a traffic quality assessment indicator set.

[0053] The traffic quality index set, the regional correlation index set, and the traffic quality assessment index set are integrated to construct a feature vector for each region.

[0054] Optionally, the system terminal first expands the scope of traffic quality assessment indicators by combining the previously extracted key traffic quality indicator set, forming a more comprehensive and detailed set of traffic quality assessment indicators. This step not only enriches the diversity of assessment indicators but also improves the accuracy and reliability of the assessment results. Subsequently, the three indicator sets are screened to remove duplicate or redundant indicators, ensuring that each indicator is unique and useful in the feature vector. Then, based on the assessment needs and objectives, it is determined which indicators are essential for constructing the feature vector and which indicators can serve as auxiliary information. It is also necessary to ensure that the data in the three indicator sets are for the same regional scope so that they can be correctly integrated. Afterward, the screened and integrated indicators are arranged in a certain order to form the basic structure of the feature vector. For each region, the corresponding indicator values ​​are filled into the corresponding positions in the feature vector. If some indicators have significant differences in scale or range between different regions, normalization can be performed to ensure they have the same weight and scale in the feature vector. Then, according to actual needs, other auxiliary information, such as the region's geographical location and demographic data, is added to the feature vector to further enrich its content. Finally, the constructed feature vector undergoes a quality check to ensure the accuracy and completeness of the data. Finally, the constructed feature vectors for each region are stored and output for subsequent analysis and application. These feature vectors not only contain traffic quality information for each region but also reflect the interactions and influences between regions, providing direction for subsequent clustering.

[0055] Based on the clustering analysis results, a road network planning model is constructed in conjunction with the road network topology diagram.

[0056] In one embodiment, after obtaining the clustering analysis results, the system terminal constructs a road network planning model in conjunction with the road network topology map. First, the system terminal interprets the clustering analysis results in detail. Clustering analysis divides massive amounts of traffic data into several traffic states or regions with distinct characteristics. By interpreting these clustering results, the differences and characteristics of different regions in terms of traffic flow, road structure, and population distribution can be understood. Subsequently, the road network topology map is analyzed in depth. This map shows the layout, connections, and hierarchical relationships of the road network, including key elements such as main roads, secondary roads, and intersections. By analyzing the topology map, bottlenecks, congestion points, and potential optimization space in the road network can be identified. After mastering the information from the clustering analysis results and the road network topology map, the system terminal begins to construct the road network planning model. This model will comprehensively consider multiple factors such as traffic demand, road structure, and regional correlation, with the goal of optimizing the road network's capacity and efficiency. During the model construction process, the system terminal employs mathematical methods or optimization algorithms, such as linear programming and integer programming, to ensure the model's accuracy and practicality. Meanwhile, the model was continuously adjusted and optimized based on actual conditions to adapt to the traffic demands and characteristics of different regions, thus constructing a comprehensive road network planning model. This model not only considers the traffic demands and characteristics of different regions but also fully takes into account the existing structure and connectivity of the road network. By optimizing the model's parameters and constraints, it can maximize the overall efficiency and operational effectiveness of the road network while meeting traffic demands. In summary, the road network planning model constructed based on cluster analysis results and the road network topology diagram can provide a scientific, reasonable, and practical road network planning solution for system terminals, helping to improve the operational efficiency and service quality of the transportation network.

[0057] The road network optimization space and the road network planning model are integrated into the multiple data processing partitions of the data processing terminal to perform road network planning and obtain multiple planning schemes;

[0058] In one embodiment, during road network planning, the system terminal first integrates the road network optimization space and road network planning model into multiple data processing partitions on the data processing end. This fully utilizes the parallel processing capabilities of the data processing end, improving the computational efficiency of road network planning. After integration, the system terminal uses these partitions to plan the road network to obtain multiple different planning schemes. Each data processing partition independently runs the road network planning model and searches for the optimal solution within the road network optimization space. Through parallel computing, multiple planning schemes can be obtained simultaneously, allowing for a more comprehensive consideration of various possible optimization strategies. This process not only improves planning efficiency but also ensures that multiple alternative schemes are obtained, facilitating subsequent comparison, analysis, and selection. In summary, by integrating the road network optimization space and road network planning model into multiple partitions on the data processing end, multiple road network planning schemes can be obtained efficiently, providing strong decision support for traffic construction and management.

[0059] Based on the simulation evaluation of the multiple planning schemes, the first road network planning scheme is obtained.

[0060] In one embodiment, the system terminal performs a simulation evaluation based on multiple previously obtained planning schemes to determine which scheme is the most ideal. Simulation evaluation is a method of testing the effectiveness of planning schemes by simulating actual traffic conditions; it helps predict the performance of each scheme in real-world operation. During the simulation evaluation, the system terminal considers various factors, including traffic flow, road capacity, vehicle speed, and congestion. By simulating the changes of these factors under different planning schemes, the advantages and disadvantages of each scheme can be understood more intuitively. Finally, after careful comparison and analysis, the first road network planning scheme is determined. This scheme demonstrates better performance in the simulation evaluation, maximizing the efficiency of the road network and reducing congestion while meeting traffic demand. In summary, by simulating and evaluating multiple planning schemes, the optimal road network planning scheme can be selected, providing a scientific basis for decision-making in actual traffic construction and management.

[0061] Furthermore, this application provides a method for obtaining a first road network planning scheme by performing simulation evaluation based on the multiple planning schemes, which further includes:

[0062] Based on the aforementioned road network topology map, and combined with urban remote sensing data, a virtual simulation city is constructed.

[0063] The multiple planning schemes are input into the virtual simulation city, run virtually, and multiple sets of scheme operation data are collected;

[0064] Based on the operational data of the multiple schemes, congestion mitigation analysis is performed to generate the first road network planning scheme.

[0065] Preferably, the system terminal constructs a highly realistic virtual simulation city based on the road network topology map and urban remote sensing data. This virtual city not only includes the layout and connection methods of roads but also incorporates detailed information such as urban topography, buildings, and green spaces, thus realistically simulating urban traffic conditions. Subsequently, multiple previously generated road network planning schemes are input into the virtual simulation city for virtual operation. During the virtual operation, various scenarios such as changes in traffic flow, vehicle travel paths, and traffic light control are simulated, and multiple sets of scheme operation data are collected. With this data, the system terminal performs congestion mitigation analysis. By comparing traffic congestion under different planning schemes, the effectiveness of each scheme in mitigating congestion is evaluated. Based on this data and analysis results, the first road network planning scheme is finally generated. This scheme performs best in the virtual simulation city, significantly reducing traffic congestion and improving the efficiency of the road network. In summary, by constructing a virtual simulation city, conducting virtual operation, and performing congestion mitigation analysis, the effectiveness of different road network planning schemes can be scientifically and objectively evaluated, thereby selecting the optimal scheme to guide actual traffic construction and management.

[0066] In summary, the embodiments of this application have at least the following technical effects:

[0067] This application embodiment uses data crawling technology to extract multi-source datasets from traffic monitoring centers and merges them with urban road network maps to obtain a road network topology map. Subsequently, point matching is performed to construct a data processing terminal, where multiple partitions process data corresponding to key traffic points. Next, a road network optimization space is set up, and data is clustered based on data correlation and traffic quality assessment indicators to obtain clustering analysis results and construct a road network planning model. The road network optimization space and planning model are integrated into the data processing terminal for road network planning, obtaining multiple planning schemes. Then, simulation evaluation is performed based on multiple planning schemes to determine the first road network planning scheme. Finally, the optimal road network planning scheme is generated through simulation evaluation. These technical effects collectively solve the technical problem that traditional road network planning methods struggle to make efficient and accurate decisions when facing complex and ever-changing traffic environments and demands, achieving intelligent road network planning based on AI and improving the operational efficiency and safety of road networks.

[0068] Example 2

[0069] Based on the same inventive concept as the AI-based road network planning method in the foregoing embodiments, such as Figure 2 As shown, this application provides a road network planning system based on AI decision-making, the system comprising:

[0070] Data fusion module 1: The data fusion module 1 is used to extract multi-source datasets from the traffic monitoring center based on data crawling technology, and to perform multi-source data fusion based on the multi-source datasets and the urban road network map to obtain a road network topology map, which contains multiple traffic points;

[0071] Point matching module 2: The point matching module 2 is used to perform point matching based on the road network topology map, and to construct a data processing terminal based on the point matching results. The data processing terminal includes multiple data processing partitions, and the multiple data processing partitions are responsible for the data processing of the corresponding traffic points.

[0072] Clustering module 3: The clustering module 3 is used to set up a road network optimization space. The road network optimization space is used to cluster data based on data correlation and traffic quality assessment indicators to obtain clustering analysis results.

[0073] Model building module 4: The model building module 4 is used to build a road network planning model based on the clustering analysis results and the road network topology diagram;

[0074] Road network planning module 5: The road network planning module 5 is used to integrate the road network optimization space and the road network planning model into the multiple data processing partitions of the data processing terminal to perform road network planning and obtain multiple planning schemes;

[0075] Simulation evaluation module 6: The simulation evaluation module 6 is used to perform simulation evaluation based on the multiple planning schemes to obtain the first road network planning scheme.

[0076] Furthermore, the data fusion module 1 is used to perform the following method:

[0077] Read the city road network map to obtain the multiple traffic points;

[0078] The multi-source dataset is split to obtain multiple split datasets, and the multiple split datasets are time-series converted to obtain multiple key point datasets;

[0079] Based on the multiple traffic points, the datasets of the multiple traffic points are matched, and the data in the same region are fused according to the data matching results to obtain the fusion results of multiple regions.

[0080] Furthermore, the data fusion module 1 is used to perform the following method:

[0081] Based on the fusion results of the multiple regions, regional accident statistical analysis is performed, and the level is assessed in combination with the regional classification standards. Based on the assessment results, multi-level nodes are determined.

[0082] Based on the regional fusion results, road attribute information is extracted, and multi-level node connections are performed based on the road attribute information to construct an initial road network topology diagram.

[0083] The distribution weights of the regional fusion results are calculated to obtain the distribution weight calculation results, which include traffic flow weights, congestion index weights, and accident frequency weights.

[0084] Based on the distribution weight calculation results, the weights are integrated, and corresponding weights are assigned to the initial road network topology diagram according to the integrated results to obtain the road network topology diagram.

[0085] Furthermore, the data fusion module 1 is used to perform the following method:

[0086] Based on the fusion results of the multiple regions, a regional traffic quality assessment is conducted to obtain the regional traffic quality assessment results.

[0087] Using AI algorithms, correlation analysis is performed on the fusion results of the multiple regions to calculate the correlation index between different regions and generate regional correlation.

[0088] Based on the regional traffic quality assessment results, the regional correlation, and the traffic quality assessment indicators, the multiple regional fusion results are clustered and divided to form multiple road network optimization spaces.

[0089] Furthermore, the data fusion module 1 is used to perform the following method:

[0090] Regional features are extracted from the fusion results of the multiple regions to obtain a first regional feature set, which includes traffic flow features, road structure features, and population distribution features.

[0091] The first region feature set is relevance index is calculated using AI algorithms, and the calculation results are standardized to obtain the first relevance index set.

[0092] The first set of relevant indicators is traversed, and corresponding weights are assigned. A weighted average is then performed to generate the regional correlation degree.

[0093] Furthermore, the data fusion module 1 is used to perform the following method:

[0094] Based on the regional traffic quality assessment results, key traffic quality indicators are extracted to form a set of key traffic quality indicators.

[0095] Based on the aforementioned regional correlation, the interactions and influences between different regions are analyzed to form a set of regional correlation indicators.

[0096] By combining the aforementioned set of key traffic quality indicators, the traffic quality assessment indicators are further expanded to form a traffic quality assessment indicator set.

[0097] The traffic quality index set, the regional correlation index set, and the traffic quality assessment index set are integrated to construct a feature vector for each region.

[0098] Furthermore, the simulation evaluation module 6 is used to perform the following method:

[0099] Based on the aforementioned road network topology map, and combined with urban remote sensing data, a virtual simulation city is constructed.

[0100] The multiple planning schemes are input into the virtual simulation city, run virtually, and multiple sets of scheme operation data are collected;

[0101] Based on the operational data of the multiple schemes, congestion mitigation analysis is performed to generate the first road network planning scheme.

[0102] It should be noted that the order of the embodiments described above is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. Furthermore, specific embodiments have been described above. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps described in the claims can be performed in a different order than that shown in the embodiments and still achieve the desired results. Additionally, the processes depicted in the drawings do not necessarily require a specific and sequential order to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0103] The above description is only a preferred embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

[0104] This specification and accompanying drawings are merely illustrative examples of this application and are intended to cover any and all modifications, variations, combinations, or equivalents within the scope of this application. Clearly, those skilled in the art can make various alterations and modifications to this application without departing from its scope. Therefore, if such modifications and modifications fall within the scope of this application and its equivalents, this application intends to include such modifications and modifications.

Claims

1. A road network planning method based on AI decision-making, characterized in that, The method includes: Based on data crawling technology, multi-source datasets are extracted from traffic monitoring centers, and multi-source data fusion is performed based on the multi-source datasets and urban road network maps to obtain a road network topology map. The road network topology map contains multiple traffic points. This step also includes: Read the city road network map to obtain the multiple traffic points; The multi-source dataset is split to obtain multiple split datasets, and the multiple split datasets are time-series converted to obtain multiple key point datasets; Based on the multiple traffic points, the datasets of the multiple traffic points are matched, and the data in the same region are fused according to the data matching results to obtain the fusion results of multiple regions. Based on the fusion results of the multiple regions, regional accident statistical analysis is performed, and a level assessment is conducted in conjunction with the regional classification standards. Based on the assessment results, multi-level nodes are determined. Based on the regional fusion results, road attribute information is extracted, and multi-level node connections are performed based on the road attribute information to construct an initial road network topology diagram. The distribution weights of the regional fusion results are calculated to obtain the distribution weight calculation results, which include traffic flow weights, congestion index weights, and accident frequency weights. Based on the distribution weight calculation results, the weights are integrated, and corresponding weights are assigned to the initial road network topology diagram according to the integration results to obtain the road network topology diagram. Based on the road network topology map, point matching is performed, and a data processing terminal is constructed based on the point matching results. The data processing terminal contains multiple data processing partitions, and the multiple data processing partitions are responsible for the data processing of the corresponding traffic points. A road network optimization space is set up, and the road network optimization space is divided into data clusters based on data correlation and traffic quality assessment indicators to obtain cluster analysis results; Based on the clustering analysis results, a road network planning model is constructed in conjunction with the road network topology diagram. The road network optimization space and the road network planning model are integrated into the multiple data processing partitions of the data processing terminal to perform road network planning and obtain multiple planning schemes; Based on the simulation evaluation of the multiple planning schemes, the first road network planning scheme is obtained.

2. The method as described in claim 1, characterized in that, A road network optimization space is established, wherein the road network optimization space is divided into data clusters based on data correlation and traffic quality assessment indicators, and the method includes: Based on the fusion results of the multiple regions, a regional traffic quality assessment is conducted to obtain the regional traffic quality assessment results. Using AI algorithms, correlation analysis is performed on the fusion results of the multiple regions to calculate the correlation index between different regions and generate regional correlation. Based on the regional traffic quality assessment results, the regional correlation, and the traffic quality assessment indicators, the multiple regional fusion results are clustered and divided to form multiple road network optimization spaces.

3. The method as described in claim 2, characterized in that, Using AI algorithms, correlation analysis is performed on the fusion results of the multiple regions to calculate correlation indicators between different regions and generate regional correlation. The method includes: Regional features are extracted from the fusion results of the multiple regions to obtain a first regional feature set, which includes traffic flow features, road structure features, and population distribution features. The first region feature set is relevance index is calculated using AI algorithms, and the calculation results are standardized to obtain the first relevance index set. The first set of relevant indicators is traversed, and corresponding weights are assigned. A weighted average is then performed to generate the regional correlation degree.

4. The method as described in claim 2, characterized in that, Before clustering the multiple regional fusion results based on the regional traffic quality assessment results, the regional correlation, and the traffic quality assessment indicators, the method further includes: Based on the regional traffic quality assessment results, key traffic quality indicators are extracted to form a set of key traffic quality indicators. Based on the aforementioned regional correlation, the interactions and influences between different regions are analyzed to form a set of regional correlation indicators. By combining the aforementioned set of key traffic quality indicators, the traffic quality assessment indicators are further expanded to form a traffic quality assessment indicator set. The traffic quality index set, the regional correlation index set, and the traffic quality assessment index set are integrated to construct a feature vector for each region.

5. The method as described in claim 1, characterized in that, Based on the multiple planning schemes, a simulation evaluation is performed to obtain a first road network planning scheme. The method includes: Based on the aforementioned road network topology map, and combined with urban remote sensing data, a virtual simulation city is constructed. The multiple planning schemes are input into the virtual simulation city, run virtually, and multiple sets of scheme operation data are collected; Based on the operational data of the multiple schemes, congestion mitigation analysis is performed to generate the first road network planning scheme.

6. A road network planning system based on AI decision-making, used to execute the road network planning method based on AI decision-making as described in any one of claims 1 to 5, characterized in that, The system includes: Data fusion module: Based on data crawling technology, it extracts multi-source datasets from the traffic monitoring center and performs multi-source data fusion with the urban road network map to obtain a road network topology map, which contains multiple traffic points; this module is also used for: Read the city road network map to obtain the multiple traffic points; The multi-source dataset is split to obtain multiple split datasets, and the multiple split datasets are time-series converted to obtain multiple key point datasets; Based on the multiple traffic points, the datasets of the multiple traffic points are matched, and the data in the same region are fused according to the data matching results to obtain the fusion results of multiple regions. Based on the fusion results of the multiple regions, regional accident statistical analysis is performed, and a level assessment is conducted in conjunction with the regional classification standards. Based on the assessment results, multi-level nodes are determined. Based on the regional fusion results, road attribute information is extracted, and multi-level node connections are performed based on the road attribute information to construct an initial road network topology diagram. The distribution weights of the regional fusion results are calculated to obtain the distribution weight calculation results, which include traffic flow weights, congestion index weights, and accident frequency weights. Based on the distribution weight calculation results, the weights are integrated, and corresponding weights are assigned to the initial road network topology diagram according to the integration results to obtain the road network topology diagram. Point matching module: performs point matching based on the road network topology map, and constructs a data processing terminal based on the point matching results. The data processing terminal contains multiple data processing partitions, which are responsible for the data processing of the corresponding traffic points. Clustering module: Set up a road network optimization space, which performs data clustering based on data correlation and traffic quality assessment indicators to obtain clustering analysis results; Model building module: Based on the clustering analysis results, a road network planning model is built in conjunction with the road network topology diagram; Road network planning module: Integrates the road network optimization space and the road network planning model into the multiple data processing partitions of the data processing terminal to perform road network planning and obtain multiple planning schemes; Simulation evaluation module: Based on the multiple planning schemes, a simulation evaluation is performed to obtain the first road network planning scheme.

Images