Main body flow determination method, electronic equipment, storage medium and program product

Abstract

The invention provides a main body flow determination method, electronic equipment, a storage medium and a program product. The subject flow determination method comprises: according to historical travel data of a road network, determining a plurality of associated paths including a target road section, the historical travel data recording a plurality of road section sets actually driven by each subject in a historical travel process, and the paths being any one road section set; according to the multi-source traffic data of the associated path, determining a main body shunt volume of the associated path in a target time period, the main body shunt volume representing the number of main bodies from a starting point area of the associated path to an end point area of the associated path along the associated path in the target time period; and taking a result obtained by adding the main branch flows of the associated paths as the main flow of the target road section in the target time period. According to the technical scheme, dependence on large-scale sample data training is avoided, and the subject flow of any road section in the road network can be accurately and efficiently estimated without a neural network.

Description

Technical Field

[0001] This disclosure relates to the field of traffic analysis technology, and more particularly to a method for determining main traffic flow, electronic equipment, storage medium, and program product. Background Technology

[0002] Determining traffic flow on road segments is fundamental to traffic state estimation and road network operation analysis, and is of great significance for applications such as traffic signal control, congestion warning, and travel guidance. Traditional methods for estimating traffic origin-end point matrices rely on parametric models and historical data calibration, and infer road segment traffic flow from traffic assignment models. However, these methods are limited by issues such as sparse observation points, strong model underdeterminism, and low computational efficiency, making it difficult to meet the needs of real-time analysis.

[0003] With the development of deep learning technology, deep learning-based traffic prediction methods have performed well in determining traffic flow on road segments. However, these methods usually rely heavily on high-quality, comprehensive training data. In real traffic systems, however, data often suffers from problems such as missing data, high noise levels, and variable scenarios, which limits the generalization ability of the model and makes it difficult to apply it stably to diverse real-world traffic scenarios. Summary of the Invention

[0004] This disclosure provides a method for determining the main flow rate, an electronic device, a storage medium, and a program product.

[0005] According to one aspect of this disclosure, a method for determining the main traffic flow is provided, comprising: determining multiple associated paths containing a target road segment based on historical travel data of a road network, wherein the historical travel data records multiple sets of road segments actually traveled by each subject during historical travel, and the path is any one of the sets of road segments; determining the main traffic flow of the associated paths in a target time period based on multi-source traffic data of the associated paths, wherein the main traffic flow represents the number of subjects traveling along the associated paths from the starting area to the ending area of ​​the associated paths in the target time period; and taking the sum of the main traffic flow of each associated path as the main traffic flow of the target road segment in the target time period.

[0006] According to one technical solution, by fusing multi-source traffic data, the associated paths containing the target road segment are identified and the main traffic flow of each path is calculated. This avoids the dependence on training with large-scale sample data and achieves accurate and efficient estimation of the main traffic flow of any road segment in the road network without the need for neural networks.

[0007] In some implementations, determining the main traffic volume distribution of the associated path during a target time period based on multi-source traffic data of the associated path includes: determining a first region where the starting point of the associated path is located and a second region where the ending point is located; multiplying the total main traffic volume in the first region during the target time period by the inter-regional departure ratio from the first region to the second region to determine the cross-regional main traffic volume from the first region to the second region during the target time period, wherein the data source providing the total main traffic volume is different from the data source providing the inter-regional departure ratio; multiplying the cross-regional main traffic volume by the selection ratio of the associated path when traveling from the first region to the second region to determine the main traffic volume distribution of the associated path when traveling from the first region to the second region during the target time period, wherein the selection ratio is determined by the historical travel data.

[0008] According to one technical solution, by integrating multi-source data from different sources, the total travel volume of the main body and the departure ratio between regions are determined respectively. Combined with the route selection ratio obtained based on historical travel data, the accurate estimation of the traffic flow of the main body of the related path is realized. While improving the accuracy of traffic prediction, it effectively avoids the bias and dependence caused by a single data source.

[0009] In some implementations, before determining the cross-regional volume of subjects from the first region to the second region during the target time period, the method further includes: expanding the offline survey data of subject travel times in the first region to determine the total travel volume of subjects whose travel times in the first region fall within the target time period.

[0010] According to one technical solution, by expanding the sample of offline survey data, the total number of main trips in the first region within the target time period can be accurately determined, which effectively improves the representativeness and accuracy of cross-regional main trip estimation, and is especially suitable for scenarios with limited actual observation data.

[0011] In some implementations, before determining the number of cross-regional entities from the first region to the second region during the target time period, the method further includes: determining, based on mobile signaling data, the number of first entities traveling during the target time period and the number of second entities traveling from the first region to the second region during the target time period; and using the ratio of the number of second entities to the number of first entities as the inter-regional departure ratio.

[0012] According to one technical solution, the departure ratio between regions is directly calculated based on mobile phone signaling data, and the intensity of traffic connections between regions is reflected by large-scale real travel trajectories, which significantly improves the objectivity and spatiotemporal resolution of cross-regional subject volume estimation.

[0013] In some implementations, before determining the main traffic flow of the associated path from the first region to the second region during the target time period, the method further includes: determining, based on the historical travel data, the number of third main entities from the first region to the second region and the number of fourth main entities along the associated path from the first region to the second region; and using the ratio of the number of fourth main entities to the number of third main entities as the selection ratio.

[0014] According to one technical solution, the route selection ratio is obtained based on historical travel data statistics, which truly reflects the route selection behavior of travelers in the actual road network and effectively improves the accuracy and reliability of traffic flow estimation for related route subjects.

[0015] In some implementations, after using the sum of the main traffic flows of each of the associated paths as the main traffic flow of the target road segment during the target time period, the method further includes: designating the road segment in the associated path equipped with a traffic monitor as another target road segment; determining the main traffic flow of the other target road segment; reading the actual traffic flow monitored by the traffic monitor for the other target road segment; and using the actual traffic flow and the main traffic flow of the other target road segment to perform feedback correction processing on the main traffic flow to obtain the corrected main traffic flow.

[0016] According to one technical solution, by introducing actual observation data from road sections equipped with traffic flow monitors, the preliminary estimated main traffic flow is corrected and adjusted, thereby realizing error feedback and parameter optimization based on real traffic conditions, which significantly improves the accuracy and reliability of main traffic flow estimation.

[0017] According to another aspect of this disclosure, an electronic device is provided, comprising: a memory storing execution instructions; and a processor executing the execution instructions stored in the memory, causing the processor to perform the main traffic determination method according to any embodiment of this disclosure.

[0018] According to another aspect of this disclosure, a readable storage medium is provided, wherein executable instructions are stored therein, which, when executed by a processor, are used to implement the main flow determination method according to any embodiment of this disclosure.

[0019] According to another aspect of this disclosure, a computer program product is provided, including a computer program that, when executed by a processor, implements the main traffic determination method described in any embodiment of this disclosure. Attached Figure Description

[0020] The accompanying drawings illustrate exemplary embodiments of the present disclosure and, together with the description thereof, serve to explain the principles of the present disclosure. These drawings are included to provide a further understanding of the present disclosure and are incorporated in and constitute a part of this specification.

[0021] Figure 1 This is a schematic diagram illustrating an application scenario of the main flow determination method according to the embodiments of this disclosure.

[0022] Figure 2 This is a flowchart of the main flow determination method according to the embodiments of this disclosure.

[0023] Figure 3 This is a schematic diagram of the main flow determination process according to the embodiments of this disclosure.

[0024] Figure 4 This is a schematic diagram of data flow according to an embodiment of the present disclosure.

[0025] Figure 5 This is a schematic block diagram of the main flow rate determination device according to an embodiment of the present disclosure.

[0026] Figure 6 This is a schematic block diagram of an electronic device according to one embodiment of the present disclosure. Detailed Implementation

[0027] The present disclosure will now be described in further detail with reference to the accompanying drawings and examples. It should be understood that the specific examples described herein are for illustrative purposes only and are not intended to limit the scope of the disclosure. Furthermore, it should be noted that, for ease of description, only the parts relevant to the present disclosure are shown in the accompanying drawings.

[0028] It should be noted that, where there is no conflict, the embodiments and features described in this disclosure can be combined with each other. The technical solutions of this disclosure will now be described in detail with reference to the accompanying drawings and embodiments.

[0029] Traffic flow is a core indicator reflecting the operational status of roads, directly affecting traffic signal control, road network capacity analysis, congestion prediction, and the formulation of traffic management strategies. In addition, in intelligent transportation systems, accurate traffic flow data is also an important basis for dynamic navigation, travel information services, traffic policy formulation, and infrastructure planning, and is of great significance for improving the level of intelligent and refined management of urban traffic.

[0030] Currently, technologies used to determine road traffic flow mainly include traditional traffic OD (origin and destination) matrix estimation methods and deep learning-based prediction methods. Traditional traffic OD matrix estimation methods rely on parametric models and iterative optimization, suffering from severe underdeterminism, low computational efficiency, and poor model generalization ability, making them unsuitable for large-scale road networks and complex traffic scenarios. While deep learning-based methods can capture complex spatiotemporal features, they heavily depend on high-quality, full-sample data input, have poor model interpretability, high computational costs, and high training and deployment barriers. Furthermore, these models have weak generalization ability across different scenarios, often requiring retraining for application transfer, further increasing the difficulty and cost of practical applications.

[0031] Therefore, this disclosure proposes a method for determining the main flow rate.

[0032] Figure 1 This is a schematic diagram illustrating an application scenario of the main traffic determination method according to the embodiments of this disclosure. For example... Figure 1 As shown, this application scenario may include a server 100 and a terminal device 200. The server 100 and the terminal device 200 can interact with each other via a network connection. The server 100 can be a cloud server or a physical server, and the terminal device 200 can be a smart device such as a computer, mobile phone, or tablet. The server 100 can receive requests from the terminal device 200 and run the subject traffic determination method disclosed herein.

[0033] Figure 2 This is a flowchart of the main flow determination method according to the embodiments of this disclosure. Figure 2 As shown, the main traffic flow determination method M200 proposes steps S210 to S230, which integrate and analyze traffic data from different channels (such as historical travel data, mobile phone signaling data, offline survey data, and actual traffic flow monitored by traffic flow monitors), cross-checking and supplementing each other, thereby improving the accuracy and reliability of main traffic flow estimation. This method not only improves estimation accuracy but also effectively reduces the problem of dependence on a large amount of sample data, enabling relatively accurate estimation of main traffic flow for road segments even in the case of scarce data, providing strong support for optimizing traffic management and alleviating road congestion.

[0034] Step S210: Based on the historical travel data of the road network, determine multiple associated paths that include the target road segment.

[0035] Historical travel data is a collection of specific road segments traversed by various traffic entities (in this disclosure, entities may include vehicles, pedestrians, and other means of transportation) during their actual travels, recorded from sources such as traffic monitoring equipment, navigation systems, satellite positioning trajectory collection equipment, or travel survey questionnaires. Each travel record should include information such as the origin, destination, travel time, and route selection, reflecting the actual route selection behavior and road network usage characteristics of traffic entities at different time periods. By analyzing this data, the relationship between various road segments and routes can be determined, including various routes related to the target route; it can also statistically analyze the proportion of route selection during trips with the same origin and destination, providing data support for determining the diversion of traffic along individual routes.

[0036] A road network refers to a transportation network structure composed of numerous nodes and road segments connected according to a specific topological relationship. It is used to describe the spatial layout and traffic relationships of road systems within a research scope, such as cities. Nodes typically represent intersections or key locations of roads, such as intersections, entrances / exits, diverging points, or merging points. The road network constitutes the basic spatial framework for traffic flow and is an important foundation for traffic planning, traffic allocation, route selection modeling, and intelligent transportation system analysis.

[0037] A path refers to an ordered combination of continuous road segments traversed from the origin to the destination in a road network. It is directional, typically composed of multiple interconnected segments, and reflects the actual travel needs of travelers and the routes they choose during their actual journeys. The "path" defined in this disclosure is not the result of all possible combinations of road segments, but rather a combination of road segments extracted from real-world travel trajectories in historical travel data, possessing both reasonableness and practical significance for travel. For example, if traffic area i is directly connected to traffic area j, and there exists a set of efficiently traversable road segments between them, but a certain combination of road segments includes unreasonable routes requiring detours to other areas, and has never been actually chosen by travelers in historical travel data, then this combination is not considered a valid path as defined in this disclosure. Therefore, the essence of a path lies in its feasibility of being chosen in reality and the authenticity of historical travel behavior.

[0038] A road segment is a road unit between two adjacent nodes in a road network. It is a basic component of a path and typically possesses attributes such as length, capacity, number of lanes, speed limit, and real-time traffic flow. A node is a location where the attributes of a road segment change, including changes in the number of lanes and speed requirements. Examples include intersections and crossroads. It is important to note that road segments have a clear directionality; a road segment pointing from one node to another and a road segment in the opposite direction are considered two independent road segments. For example, a road segment pointing from crossroads 1 to crossroads 2 and a road segment returning from crossroads 2 to crossroads 1 are considered two different road segments in the road network, corresponding to different travel directions and traffic flow characteristics.

[0039] The target road segment refers to the road unit whose main traffic flow (such as vehicle flow) needs to be determined, and it is the research object of traffic flow analysis in this method. In practical applications, any road segment in the road network can be used as the target road segment. By analyzing the main traffic flow of each road segment, traffic congestion points or operational bottlenecks in the road network can be identified. Associated paths are all paths in the road network that pass through the target road segment. The starting and ending points of each associated path can be different, and there are no restrictions here.

[0040] Step S220: Determine the main traffic flow distribution of the associated path during the target time period based on the multi-source traffic data of the associated path.

[0041] Multi-source traffic data encompasses a wide range of information sources, including historical travel data, mobile signaling data, and offline survey data. When calibrating the estimated main traffic flow distribution, it also includes actual traffic flow data from road segments obtained from traffic monitors. Historical travel data provides actual driving records for each path during historical trips, clarifying the selectable paths and the selection ratio of each path when the origin and destination are the same. Mobile signaling data tracks the movement trajectory of mobile phone users, indirectly linking it to the entity bound to the phone (e.g., the user or the vehicle driven by the user), used to clarify the proportion of cross-regional departures of the entity, serving as process data for determining the main traffic flow distribution. Offline survey data captures travelers' travel habits and preferences, providing a sample basis for determining the total travel volume of the entity within a certain time period; the total travel volume of the entity is process data for determining the main traffic flow distribution. Through multi-source traffic data, the main traffic flow distribution of associated paths within a specific time period can be analyzed and determined more comprehensively and accurately, improving the estimation accuracy of the main traffic flow distribution and avoiding the interference of bias from a single data source on the estimation results.

[0042] Main traffic flow distribution represents the number of main traffic sources along an associated path from its starting area to its ending area during a target time period. It characterizes the actual traffic demand and flow distribution along that path. Specifically, main traffic flow distribution not only reflects the frequency with which travelers choose that path within a specific time period but also demonstrates the path's importance within the overall road network. By statistically analyzing the main traffic flow distribution of each associated path, we can clearly understand the distribution of traffic sources and destinations upstream and downstream of the target road segment, providing crucial information for subsequent main traffic flow estimation, traffic flow allocation optimization, and congestion cause analysis.

[0043] Step S230: The sum of the main traffic flows of each associated path is taken as the main traffic flow of the target road segment during the target time period.

[0044] In traffic flow analysis, the target time period refers to the selected time interval for assessing the main traffic flow on a specific road segment or route. This time period can be flexibly set according to research objectives or actual needs, and can be peak hours, off-peak hours, a fixed time interval on weekdays (such as every hour), weekends, or holidays—all representative periods. The selection of the target time period is crucial for accurately understanding traffic dynamics and patterns, as it directly affects the effectiveness and relevance of data collection and analysis. For example, when conducting traffic congestion analysis, weekday morning and evening rush hours might be chosen as the target time period; while when assessing nighttime road use, later evening hours would be selected. By conducting main traffic flow analysis for different target time periods, traffic management departments can identify traffic characteristics and patterns within different time periods, thereby developing more scientific and reasonable management and control strategies.

[0045] Mainstream traffic flow refers to the statistical data of the main traffic passing through a target road segment within a specific time period. It is a key indicator for measuring the road segment's capacity and traffic operation status. Mainstream traffic flow not only reflects the actual usage of each road segment in the road network, but also reveals important information such as traffic congestion, traffic efficiency, and the distribution of travel demand.

[0046] Since the main traffic flow of the target road segment originates from all associated paths passing through it, the total traffic demand on the target road segment can be accurately reconstructed by summing the main traffic flows of each associated path. This method fully leverages the advantages of multi-source traffic data and combines it with actual path selection behavior to improve the accuracy and practicality of road segment traffic estimation, providing reliable data support for traffic management, congestion analysis, and road network optimization.

[0047] Figure 3 This is a schematic diagram of the main flow determination process according to the embodiments of this disclosure. The following is in conjunction with... Figure 3 The process of determining the main flow rate is explained.

[0048] In step 301, the total number of trips in the first region is determined.

[0049] In determining the main traffic flow distribution of the path to be analyzed, the area where the starting point of the path is located is defined as the first area, and the area where the ending point of the path is located is defined as the second area. Here, "area" refers to a basic transportation unit within the road network, defined based on its geographical spatial layout or traffic function characteristics, possessing a certain capacity for generating or attracting travel. For example, residential areas, commercial centers, industrial parks, and transportation hubs are all typical areas with relatively stable travel characteristics. These areas typically serve as the origin and destination points for travel, and are used in traffic planning and analysis to summarize and statistically analyze travel generation and attraction, forming an important foundation for characterizing travel distribution patterns.

[0050] In practical applications, the entire road network is typically divided into multiple independent and non-overlapping regions. Each region contains several road segments that converge at a representative central point, known as the "centroid," which is usually considered the origin and destination of trips within that region. The geographical extent and the number of road segments contained in each region can vary, and this method does not impose strict restrictions on this.

[0051] After clarifying the concept of a region, mobile signaling data can be used as one data source to obtain the total travel volume of the main body in the first region; alternatively, offline survey data can be used as another data source. When using offline survey data as the data source for extracting the total travel volume, the offline survey data needs to be expanded. That is, based on the collection scope and travel characteristics of the survey sample, combined with information such as the resident population, age structure, and travel rate of the first region, the sample data can be expanded to the scope of the first region using weighted amplification or proportional extrapolation to obtain a representative total travel volume.

[0052] In addition, mobile signaling data, provided by operators, consists of signaling records about user location, including base station handover and location updates. This data can identify the user's origin and destination areas within a target time period. By performing spatiotemporal matching and dwell time determination on the signaling data, valid travel records are filtered out. Combined with user sampling ratios, an overall extrapolation can be performed to obtain the total number of trips in that area.

[0053] It should be noted that when determining the overall state of the road network, all road segments should be treated as target road segments to determine the main traffic flow of all road segments. Therefore, the first region here can be the region where the starting point of the associated path of the target road segment is located.

[0054] In step 302, the departure ratio between the first area and the second area is determined.

[0055] The inter-regional departure ratio represents the proportion of entities traveling from one traffic area to another relative to the total number of entities traveling from the originating area. It characterizes the distribution of entities within a given area across different regions, reflecting the strength of inter-regional travel connections and the direction of traffic flow. This data is determined from mobile phone signaling data. Specifically, based on the location information of mobile phone users at different times, the originating and destination areas of travel within the target time period are identified. Then, the number of entities departing from the originating area to the destination area during the target time period is counted, along with the total number of entities traveling from the originating area during the target time period. Finally, the ratio of the number of entities departing from each area to the total number of entities traveling from the originating area is used as the inter-regional departure ratio.

[0056] In this disclosure, the area where the starting point of the associated path is located is designated as the first area, and the area where the ending point of the associated path is located is designated as the second area. Therefore, the interval departure ratio is the ratio of the number of main departures from the first area to the second area to the total number of main trips in the first area.

[0057] In step 303, the cross-regional entity quantity from the first region to the second region is determined.

[0058] Cross-regional traffic volume is the result of multiplying the total traffic volume of main entities by the inter-regional departure ratio. It should be noted that, to improve the accuracy of the estimation, expanded sample results from offline surveys are used as the data source for the total traffic volume of main entities in the first region; mobile signaling data is used as the data source for the inter-regional departure ratio. Compared to estimation methods using a single data source, the multi-source data fusion strategy effectively avoids the inherent coverage differences and sample limitations of each data source, improves the reliability of cross-regional traffic volume estimation, and provides more accurate data support for subsequent determination of main traffic flow on road segments.

[0059] In step 304, the main traffic flow of the associated path from the first region to the second region is determined based on the selection ratio of the associated path.

[0060] The selection ratio is determined based on historical travel data, representing the proportion of main vehicles traveling along a certain route when they share the same origin and destination areas. Step 303 determines the cross-regional vehicle volume along the associated route from the first area to the second area. Then, multiplying the cross-regional vehicle volume by the main vehicle flow rate along the associated route yields the number of main vehicles choosing the associated route during the journey from the first area to the second area, i.e., the main vehicle flow rate.

[0061] It should be noted that historical travel data was used as the data source for determining the selection ratio in this process; while the cross-regional subject volume was determined by offline survey data and mobile signaling data. The combination of the three data sources further avoids the errors and limitations of a single data source.

[0062] In step 305, the main flow rate is corrected by using the actual flow rate of the monitored road segments.

[0063] In principle, the main traffic flow of a target road segment can be obtained by adding the main traffic flows of all its associated paths. However, due to the cumulative effect of errors in path traffic estimation, the result obtained by direct addition may deviate from the true value. To address this, this disclosure introduces a feedback correction mechanism, which adjusts the main traffic flow by incorporating real observation data (actual traffic flow of the road segment read by a traffic monitor), thereby achieving a refined correction of the main traffic flow.

[0064] Specifically, the road segment with a traffic monitor in the associated path is designated as another target road segment; the main traffic flow of the other target road segment is determined; the actual traffic flow of the other target road segment monitored by the traffic monitor is read; and the main traffic flow of the other target road segment is corrected by feedback using the actual traffic flow and the main traffic flow of the other target road segment to obtain the corrected main traffic flow.

[0065] In real-world traffic scenarios, some key road sections are equipped with traffic flow monitoring devices (such as geomagnetic sensors, cameras, and radar). These devices can record the actual traffic flow on that road section during a target time period. Although not all road sections in the network have such monitoring capabilities, existing observational data can serve as a reference standard for error correction. By constructing multi-level error functions, the deviations between key variables such as total trip volume, inter-regional departure ratio, and main traffic flow and their actual values ​​can be quantified. Furthermore, the chain rule is used to progressively correct these errors, starting from determining the total trip volume and working its way up to determining the main traffic flow, thus correcting the main traffic flow on road sections within the same traffic area that are not equipped with traffic flow monitoring.

[0066] Specifically, the model adopts a weighted loss function (LOSS) to comprehensively consider the total travel volume error. Inter-regional departure ratio error Main flow error Actual flow error And the target road segment speed error used to calibrate actual traffic flow data. And by introducing , , , and As respectively to The weighting coefficients are used to adjust the priority of each error term in the optimization process. This mechanism can not only effectively integrate multi-source real data, but also dynamically adjust according to the credibility differences of different data types. For example, when the expansion results of survey data or mobile signaling data are poor, their credibility is reduced. This ensures the overall travel trend while shifting the focus of verification towards the survey volume of the target road segment, thereby minimizing global loss while ensuring physical rationality.

[0067] More specifically, a multi-level loss function (LOSS) is constructed to evaluate errors in dimensions such as total trip volume, inter-regional departure ratio, main traffic flow, and actual traffic flow. The formula is shown below.

[0068] ,in to Used to adjust the contribution of each error term to the overall optimization, since the actual main flow is the monitored real flow, it has higher reliability, and its weighting coefficient... It should have a larger value.

[0069] ,in The number of traffic zones in the road network. Signs indicating traffic areas Indicates the first The actual total number of trips by the main entities in each transportation area Indicates the first The total number of trips in each transportation area is reduced by gradually decreasing the loss. Gradually getting closer .

[0070] ,in Indicates by the first The traffic zone to the first Total number of routes in each traffic area Indicates from the first The traffic zone to the first Inter-regional departure ratio of each traffic zone Indicates from the first The traffic zone to the first The actual inter-region departure ratio of each traffic zone is achieved by gradually reducing the loss ratio. Gradually getting closer .

[0071] ,in, This indicates the total number of road segments in the road network. For road segment markings, For road section The main flow, For road section The actual main traffic flow. Among them, road sections It refers to the road segment in the associated path that has a traffic monitor.

[0072] ,in, This indicates the total number of road segments in the road network equipped with traffic flow monitors. To configure the road segment identification for traffic flow monitors, For road section The main flow, For road section The actual main flow.

[0073] ,in, For road section The main velocity, For road section The actual speed of the main body.

[0074] During feedback correction, the gradient of the total loss function with respect to all trainable parameters is calculated using the chain rule. That is, the gradient of the total loss with respect to the parameters is a weighted sum of the gradients of each sub-loss term, reflecting the collaborative optimization mechanism of errors in multi-task learning. The partial derivatives of each sub-loss term with respect to the output variable represent the deviation between the estimated and true values.

[0075] These local gradients will be corrected through automatic differentiation, ultimately guiding the direction of parameter updates.

[0076] The gradient of the total loss function is: in for , and Any one of them.

[0077] The gradients of the sub-loss terms are as follows: , , , , .

[0078] During parameter optimization, the Adam (Adaptive Moment Estimation) optimizer can be used. Based on the adaptive learning rate mechanism of first- and second-order momentum, it can efficiently update parameters such as total traffic volume, traffic flow error, and traffic flow distribution. This method ensures the convergence speed of traffic flow and makes the estimation results closer to the actual traffic flow status of road segments.

[0079] ,in Indicates the current iteration process , or The value, Indicates the previous iteration , or The value of . For the number of iterations, This is a first-moment estimate of the momentum term's representative parameter; The variance term represents the second moment estimate of the parameter; Indicates the learning rate for different layers; This represents the parameter smoothing term. Wherein, , , , These are all attribute values ​​of the optimizer.

[0080] By introducing a feedback correction mechanism and combining it with existing observation data for error feedback and parameter adjustment, the accuracy of the main traffic flow estimation of the target road segment can be effectively improved in complex road networks lacking comprehensive monitoring. This makes the final output of "corrected main traffic flow" closer to the actual traffic operation, and has higher practical value and engineering feasibility.

[0081] In step 306, it is determined whether the number of iterations for each parameter during the feedback correction process exceeds a threshold. The threshold is one of the standards for measuring the effectiveness of iteration; the number of iterations can be set as needed and is not limited here.

[0082] When the number of iterations is less than or equal to the threshold, step 305 is executed repeatedly until the number of iterations is greater than the threshold, triggering step 307, where the main flow after iteration is used as the corrected main flow.

[0083] Figure 4 This is a schematic diagram of data flow according to an embodiment of this disclosure. For example... Figure 4 As shown, this illustrates the flow and relationships of data when determining the main traffic volume of a road segment.

[0084] Specifically, with the first Taking a specific traffic area as an example, and using offline survey data as the data source, the first [item] in the target time period was determined. Total travel volume of each transportation area Then, using mobile signaling data as the data source, the first [item] in the target time period is determined. The number of the first entities in each transportation zone, and from the... The traffic zone to the first The number of the second entity in each traffic area is used as the ratio of the number of the second entity to the number of the first entity, and then the ratio of the number of the second entity to the number of the first entity is used as the number of the second entity. The traffic zone to the first Inter-regional departure ratio of each traffic zone .

[0085] Furthermore, the total number of main trips in the first region during the target period is multiplied by the inter-regional departure ratio from the first region to the second region to determine the cross-regional main trip volume from the first region to the second region during the target period. ,in .

[0086] Using historical travel data as the data source, determine the first The traffic zone to the first The traffic area along the first n The proportion of each path selected Specifically, based on historical travel data, the number of third entities traveling from the first region to the second region during historical travel processes was determined, as well as the number of third entities with the highest number of third entities being the [number missing]. n Each path is considered as the fourth element of the travel route; the ratio of the fourth element to the third element is used as the selection ratio. .

[0087] Furthermore, the cross-regional volume of entities is multiplied by the proportion of entities choosing associated paths when traveling from the first region to the second region to determine the entity distribution flow of associated paths when traveling from the first region to the second region during the target time period. That is... .

[0088] It should be noted that to understand the traffic flow of all road segments in the entire road network, it is necessary to determine the total main trip volume of all traffic areas; for each traffic area, it is also necessary to trace the cross-regional main trip volume between it and other traffic areas, as well as the main trip volume of each path between every two traffic areas. This will allow for an exhaustive understanding of the main trip volume of all road segments.

[0089] In other words, it is necessary to determine the relationship between paths and road segments, i.e., the set of road segments corresponding to a path, based on historical travel data of the road network. Then, a two-dimensional sparse matrix is ​​constructed between paths and road segments. .in, This represents the total number of paths in the road network. This represents the total number of all road segments in the road network. The elements in the matrix... This element indicates that the path contains the road segment. This indicates that the path does not contain this element.

[0090] Furthermore, determine the main traffic flow for each road segment. , where the symbol This is the transpose symbol. Indicates the first The main traffic flow of the first road section is the first The sum of the main traffic flows of each associated path in each road segment. It is the collection of the main traffic flows of all road segments in the road network. to These are the main traffic flows distributed to each path in the road network, among which... This represents the main traffic distribution of the first path. This represents the main flow distribution of the p-th path.

[0091] Finally, the main vehicle speed for each road segment is determined based on the main traffic flow. The main traffic flow is then used as input data and fed into a well-known model (such as the classic flow density-velocity model, Greenhill, Greenberg, Underwood models, etc.) to calculate the speed value. ,in This represents the velocity calculation model.

[0092] Figure 5 This is a schematic block diagram of the main flow rate determination device according to an embodiment of this disclosure. Figure 5 As shown, this disclosure proposes a main traffic flow determination device 500, comprising: an associated path determination module 510, used to determine multiple associated paths containing a target road segment based on historical travel data of the road network, wherein the historical travel data records multiple sets of road segments actually traveled by each subject during historical travel, and the path is any set of road segments; a first analysis module 520, used to determine the main traffic flow of the associated path in a target time period based on multi-source traffic data of the associated path, wherein the main traffic flow represents the number of subjects along the associated path from the starting area to the ending area of ​​the associated path in the target time period; and a second analysis module 530, used to take the sum of the main traffic flow of each associated path as the main traffic flow of the target road segment in the target time period.

[0093] The main flow determination device 500 disclosed herein can be in the form of computer software, and each module of the main flow determination device 500 can be in the form of computer software modules.

[0094] The various modules of the main flow determination device 500 disclosed herein are set up to implement the various steps of the main flow determination method. The execution principle and steps can be referred to the above text and will not be repeated here.

[0095] Figure 6 This is a schematic block diagram of an electronic device according to one embodiment of the present disclosure. Figure 6 As shown, this disclosure also provides an electronic device 1000, including: a processor 1200 and a memory 1300, the memory 1300 storing execution instructions; the processor 1200 executes the execution instructions stored in the memory 1300, causing the processor 1200 to execute a main flow determination method.

[0096] The hardware architecture of the electronic device 1000 can be implemented using a bus architecture. The bus architecture can include any number of interconnect buses and bridges, depending on the specific application of the hardware and overall design constraints. Bus 1100 connects various circuits, including one or more processors 1200, memory 1300, and / or hardware modules. Bus 1100 can also connect various other circuits 1400, such as peripheral devices, voltage regulators, power management circuits, external antennas, etc.

[0097] Bus 1100 can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Component (EISA) bus, etc. Buses can be categorized as address buses, data buses, control buses, etc. For ease of representation, only one connection line is used in this diagram, but this does not imply that there is only one bus or only one type of bus.

[0098] This disclosure also provides a readable storage medium storing a computer program that, when executed by a processor, is used to implement the methods described above. A "readable storage medium" can be any means capable of containing, storing, communicating, propagating, or transmitting a program for use by or in conjunction with an instruction execution system, apparatus, or device. More specific examples of a readable storage medium include: an electrical connection with one or more wires (electronic device), a portable computer disk drive (magnetic device), random access memory (RAM), read-only memory (ROM), erasable and programmable read-only memory (EPROM or flash memory), fiber optic devices, and portable read-only memory (CDROM), etc.

[0099] This disclosure also provides a computer program product, the methods of which can be implemented wholly or partially through software, hardware, firmware, or any combination thereof. When implemented in software, it can be implemented wholly or partially as a computer program product. The computer program product includes one or more computer programs or instructions. When the computer program or instructions are loaded and executed, all or part of the processes or functions of this disclosure are performed.

[0100] Computer programs or instructions can be stored in a readable storage medium or transferred from one readable storage medium to another. For example, the computer program or instructions can be transferred from one website, computer, server, or data center to another website, computer, server, or data center via wired or wireless means. The readable storage medium can be any available medium capable of access, or a data storage device such as a server or data center that integrates one or more available media. The available medium can be a magnetic medium, such as a floppy disk, hard disk, or magnetic tape; an optical medium, such as a digital video optical disc; or a semiconductor medium, such as a solid-state drive. The computer-readable storage medium can be a volatile or non-volatile storage medium, or it can include both volatile and non-volatile types of storage media.

[0101] Those skilled in the art will understand that embodiments of this disclosure can be provided as methods, electronic devices, readable storage media, or computer program products. Therefore, this disclosure can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this disclosure can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0102] This disclosure is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to this disclosure. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0103] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0104] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0105] Those skilled in the art should understand that the above embodiments are merely for illustrating the present disclosure and are not intended to limit the scope of the disclosure. Those skilled in the art can make other changes or modifications based on the above disclosure, and these changes or modifications still fall within the scope of the present disclosure.

[0106] It is understood that before using the technical solutions disclosed in the various embodiments of this disclosure, users should be informed of the types, scope of use, and usage scenarios of the personal information involved in this disclosure in an appropriate manner in accordance with relevant laws and regulations, and user authorization should be obtained.

[0107] For example, upon receiving a user's active request, a prompt message is sent to the user to explicitly inform them that the requested operation will require the acquisition and use of the user's personal information. This allows the user to independently choose whether to provide personal information to the software or hardware, such as the electronic device, application, server, or storage medium performing the operations of this disclosed technical solution, based on the prompt message.

[0108] As an optional but non-limiting implementation, in response to a user's active request, sending a prompt message to the user can be done via a pop-up window, where the prompt message can be presented in text format. Furthermore, the pop-up window can also include a selection control allowing the user to choose "agree" or "disagree" to provide personal information to the electronic device.

[0109] It is understood that the above notification and user authorization process are merely illustrative and do not constitute a limitation on the implementation of this disclosure. Other methods that comply with relevant laws and regulations may also be applied to the implementation of this disclosure.

[0110] At the same time, it is understood that the data involved in this disclosed technical solution (including but not limited to the data itself, the acquisition or use of the data) shall comply with the requirements of relevant laws, regulations and related provisions.

[0111] In the description of this specification, the references to terms such as "one embodiment / mode," "some embodiments / modes," "example," "specific example," or "some examples," etc., refer to specific features, structures, or characteristics described in connection with that embodiment / mode or example, which are included in at least one embodiment / mode or example of this disclosure. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment / mode or example. Moreover, the specific features, structures, or characteristics described may be combined in any suitable manner in one or more embodiments / modes or examples. Furthermore, without contradiction, those skilled in the art can combine and integrate the different embodiments / modes or examples described in this specification, as well as the features of different embodiments / modes or examples.

[0112] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this disclosure, "a plurality of" means at least two, such as two, three, etc., unless otherwise explicitly specified.

Claims

1. A method for determining the main flow rate, characterized in that, include: Based on historical travel data of the road network, multiple associated paths containing the target road segment are determined. The historical travel data records the various road segment sets that each subject actually traveled in the historical travel process, and the path is any one of the road segment sets. Based on the multi-source traffic data of the associated path, the main traffic flow of the associated path in the target time period is determined. The main traffic flow represents the number of main traffic items along the associated path from the starting area to the ending area of ​​the associated path in the target time period. as well as The sum of the main traffic flows of each associated path is taken as the main traffic flow of the target road segment during the target time period.

2. The method for determining the main flow rate according to claim 1, characterized in that, Based on the multi-source traffic data of the associated path, determine the main traffic flow distribution of the associated path in the target time period, including: Determine the first region where the starting point of the associated path is located and the second region where the ending point is located; The total number of main trips in the first region during the target time period is multiplied by the inter-regional departure ratio from the first region to the second region to determine the cross-regional main trip volume from the first region to the second region during the target time period, wherein the data source providing the total number of main trips is different from the data source providing the inter-regional departure ratio; The cross-regional volume is multiplied by the selection ratio of the associated path when traveling from the first region to the second region to determine the main flow of the associated path when traveling from the first region to the second region during the target time period, wherein the selection ratio is determined by the historical travel data.

3. The method for determining the main flow rate according to claim 2, characterized in that, Before determining the cross-regional volume of entities from the first region to the second region during the target time period, the method further includes: The offline survey data on the travel time of the main subjects in the first region are expanded to determine the total number of main subjects whose travel time in the first region falls within the target time period.

4. The method for determining the main flow rate according to claim 3, characterized in that, Before determining the cross-regional volume of entities from the first region to the second region during the target time period, the method further includes: Based on mobile signaling data, determine the number of first entities traveling during the target time period, and the number of second entities traveling from the first area and ending at the second area during the target time period; and The ratio of the number of the second entity to the number of the first entity is used as the departure ratio between the regions.

5. The method for determining the main flow rate according to claim 2, characterized in that, Before determining the main traffic distribution of the associated path from the first region to the second region during the target time period, the method further includes: Based on the historical travel data, determine the number of third entities traveling from the first region to the second region, and the number of fourth entities traveling from the first region to the second region along the associated path; and The ratio of the number of the fourth subject to the number of the third subject is used as the selection ratio.

6. The method for determining the main flow rate according to claim 1, characterized in that, After summing the main traffic flows of each associated path as the main traffic flow of the target road segment during the target time period, the method further includes: The road segment in the associated path that is configured with a traffic monitor is designated as another target road segment; Determine the main traffic flow of the other target road segment; Read the actual traffic flow monitored by the traffic flow monitor for the other target road segment; as well as Using the actual traffic flow and the main traffic flow of the other target road segment, the main traffic flow is subjected to feedback correction processing to obtain the corrected main traffic flow.

7. An electronic device, characterized in that, include: The memory stores execution instructions; as well as A processor that executes the execution instructions stored in the memory, causing the processor to perform the main flow determination method according to any one of claims 1 to 6.

8. A readable storage medium, characterized in that, The readable storage medium stores execution instructions, which, when executed by a processor, are used to implement the main flow determination method according to any one of claims 1 to 6.

9. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by the processor, it implements the main flow determination method according to any one of claims 1 to 6.

Images