Road network partitioning method and device, computer equipment and computer readable storage medium

Abstract

The invention discloses a road network partitioning method and device, computer equipment and a computer readable storage medium, and relates to the technical field of computers, real-time distribution scene data are fused, so that the characteristics of both natural segmentation and real-time distribution scenes are considered in the road network partitioning process, and meanwhile, idle road network data are eliminated, so that the real-time distribution scene data is obtained. The scale of the road network to be partitioned is reduced, the partitioning efficiency is improved, the reasonability and the flexibility are good, and then the accuracy and the high efficiency of subsequent path planning are guaranteed. The method comprises the following steps: determining to-be-partitioned road network data, and removing idle road network data from the to-be-partitioned road network data to obtain preprocessed road network data; obtaining instant distribution scene data matched with the road network data to be partitioned, and setting a road section attribute for each road section included in the preprocessed road network data according to the instant distribution scene data; and according to the road section attribute corresponding to each road section included in the preprocessed road network data, performing partitioning processing on the preprocessed road network data to obtain a road network partitioning result.

Description

Technical Field

[0001] The present application relates to the field of computer technology, and in particular to a road network partitioning method, apparatus, computer equipment, and computer-readable storage medium. Background Art

[0002] In the field of instant delivery services, high-quality route planning services are the foundation for efficient and accurate delivery. Route planning must not only consider the shortest distance in geographic space, but also comprehensively consider multiple factors such as real-time traffic conditions, road properties, delivery range, business district distribution, and points of interest (AOIs) to ensure timely delivery and user experience. The implementation of route planning algorithms often relies on efficient road network partitioning technology to enable rapid processing of large-scale road network data and route calculation.

[0003] In related technologies, road network zoning is primarily based on natural features such as rivers and mountains, combined with planned area data for multi-level segmentation. However, the applicant recognized that in the context of instant delivery, considering only natural geographical features would result in a road network zoning result that is inconsistent with the instant delivery scenario. Furthermore, the large scale of the road network in this context can easily exacerbate the performance bottleneck of road network zoning, resulting in poor rationality and flexibility in road network zoning, making subsequent route planning less accurate and efficient. Summary of the Invention

[0004] In view of this, the present application provides a road network partitioning method, apparatus, computer equipment and computer-readable storage medium, the main purpose of which is to solve the problem that in the current instant delivery scenario, only considering natural geographical features will lead to road network partitioning results that are inconsistent with the instant delivery scenario, and the road network scale in the instant delivery scenario is large, which easily aggravates the performance bottleneck of road network partitioning, and the rationality and flexibility of road network partitioning are poor, making subsequent path planning not accurate and efficient enough.

[0005] According to a first aspect of the present application, a road network partitioning method is provided, the method comprising:

[0006] Determining road network data to be partitioned, removing idle road network data from the road network data to be partitioned, and obtaining preprocessed road network data;

[0007] Acquire instant delivery scenario data that matches the road network data to be partitioned, and set a road segment attribute for each road segment included in the pre-processed road network data according to the instant delivery scenario data;

[0008] According to the road segment attributes corresponding to each road segment included in the pre-processed road network data, the pre-processed road network data is partitioned to obtain a road network partition result.

[0009] Optionally, removing idle road network data from the road network data to be partitioned to obtain pre-processed road network data includes:

[0010] Matching the road network data to be partitioned with historical order data within a specified historical time period, aggregating delivery trajectory data in the historical order data, and determining at least one delivery activity area in the road network data to be partitioned;

[0011] Determining a preset expansion distance, expanding the range of each of the delivery activity areas according to the expansion distance, and using at least one expanded delivery activity area as at least one common road network area;

[0012] The road network data to be partitioned, which are located between the at least one commonly used road network area, are used as the idle road network data, and the idle road network data are eliminated, and the eliminated road network data to be partitioned are used as the preprocessed road network data.

[0013] Optionally, the acquiring of instant delivery scenario data and setting a road segment attribute for each road segment included in the pre-processed road network data according to the instant delivery scenario data include:

[0014] Acquire the instant delivery scenario data that matches the road network data to be partitioned, the instant delivery scenario data including but not limited to business district range data, delivery range data, and point of interest area AOI boundary data;

[0015] Performing spatial intersection processing on the instant delivery scenario data and the pre-processed road network data, determining at least one first road segment intersecting with the instant delivery scenario data from among the road segments included in the pre-processed road network data, and setting a road segment attribute for the at least one first road segment using a preset intersection identifier;

[0016] Determining at least one second road segment in the pre-processed road network data, and setting a road segment attribute for the at least one second road segment using a preset non-intersection identifier, wherein the at least one second road segment is a road segment other than the at least one first road segment among the road segments included in the pre-processed road network data;

[0017] According to the instant delivery scenario data, counting the road section flow data of each road section included in the pre-processed road network data, and determining the road section type data of each road section;

[0018] Based on the road segment flow data and the road segment type data of each road segment, the road segment attributes currently set for the road segments included in the pre-processed road network data are adjusted to complete the setting of the road segment attributes.

[0019] Optionally, adjusting the road segment attributes currently set for the road segments included in the pre-processed road network data based on the road segment flow data and the road segment type data of each road segment includes:

[0020] Determining at least one road section to be adjusted from the road sections included in the pre-processed road network data, and reading the road section attributes currently set for each of the road sections to be adjusted, wherein the road section flow data corresponding to the at least one road section to be adjusted is lower than a preset flow threshold and the corresponding road section type data belongs to a preset special road section type;

[0021] Determining, among the at least one road section to be adjusted, a first road section to be adjusted whose road section attribute is currently set to the intersection identifier, and adjusting the road section attribute of the first road section to be adjusted to a preset special identifier;

[0022] A second road section to be adjusted whose road section attribute is currently set to the non-intersection mark is determined in the at least one road section to be adjusted, and the road section attribute of the second road section to be adjusted is adjusted to the intersection mark.

[0023] Optionally, performing partition processing on the pre-processed road network data according to the road segment attributes corresponding to each road segment included in the pre-processed road network data to obtain a road network partition result includes:

[0024] Performing local road network merging on the pre-processed road network data to obtain pre-segmented road network data, and obtaining a preset weight setting strategy, respectively querying the weight value corresponding to the section attribute of each road section in the preset weight setting strategy, setting a section weight for each road section according to the queried weight value, adopting a preset section segmentation algorithm, segmenting the pre-segmented road network data according to the section weight corresponding to each road section, obtaining the pre-segmented road network data with segmentation results, and restoring the pre-segmented road network data with segmentation results to the pre-processed road network data with segmentation results, and using the pre-processed road network data with segmentation results as the road network partitioning result; or,

[0025] Read the historical road network partitioning results of the road network data to be partitioned, extract the historical pre-processed road network data with the segmentation results with the smallest time interval between the segmentation time point and the current time point from the historical road network partitioning results, refer to the section attributes corresponding to each road section included in the pre-processed road network data, perform difference marking on the pre-processed road network data and the historical pre-processed road network data with the segmentation results to obtain a difference marking result, and partition the pre-processed road network data according to the difference marking result to obtain the road network partitioning result.

[0026] Optionally, the performing local road network merging on the pre-processed road network data to obtain pre-segmented road network data includes:

[0027] Randomly selecting any road section node from the road section nodes included in the pre-processed road network data as a target road section node, performing a breadth-first search with the target road section node as a starting point, and when a preset number of designated road section nodes are found, using the road network formed by the preset number of designated road section nodes as a sub-road network;

[0028] Sorting the preset number of designated road section nodes according to the order in which they are searched, and dividing the sorted preset number of designated road section nodes into a source node group and a sink node group according to a preset division ratio;

[0029] Dividing the sub-road network into two temporary road networks using a maximum flow minimum cut algorithm according to the source node group and the sink node group, taking the temporary road network including the target road segment node of the two temporary road networks as a local road network, and taking the road segment connected to the local road network in the pre-processed road network data as a boundary segment of the local road network;

[0030] When the local road network meets preset merging conditions, average node coordinates of all road segment nodes included in the local road network are calculated, the calculated average coordinates are used as temporary node coordinates, and a temporary road segment node is created at the temporary node coordinates, wherein the preset merging conditions include but are not limited to the density of road network nodes and the parameter threshold of the maximum flow minimum cut algorithm;

[0031] Establishing a temporary road section between the temporary road section node and the boundary road section, and deleting all road section nodes and all road sections included in the local road network from the pre-processed road network data, thereby completing a merging operation on the local road network;

[0032] A road segment node is randomly selected from other road segment nodes included in the preprocessed road network data as a new target road segment node, a breadth-first search is performed again with the new target road segment node as the starting point, and a new local road network is re-determined, and a merging operation is performed on the new local road network until there are no selectable road segment nodes in the preprocessed road network data, completing the operation of merging the local road network of the preprocessed road network data to obtain the pre-segmented road network data, wherein the other road segment nodes are the road segment nodes in the preprocessed road network data except the target road segment node and the preset number of designated road segment nodes.

[0033] Optionally, the restoring the pre-segmented road network data with segmentation results to the pre-processed road network data with segmentation results includes:

[0034] Determining at least one temporary road segment node in the pre-segmented road network data with the segmentation result, and querying the pre-processed road network data for a local road network corresponding to each temporary road segment node, wherein the node coordinates of the temporary road segment node are calculated by averaging the node coordinates of all road segment nodes included in the local road network corresponding to the temporary road segment node;

[0035] The local road network corresponding to each temporary road section node is re-added to the pre-segmented road network data with the segmentation results, and each temporary road section node and the temporary road section established based on each temporary road section node are deleted, and the pre-segmented road network data with the segmentation results after the addition and deletion processing is used as the preprocessed road network data with the segmentation results.

[0036] Optionally, referring to the road segment attributes corresponding to each road segment included in the preprocessed road network data, performing difference marking on the preprocessed road network data and the historical preprocessed road network data with the segmentation result, to obtain a difference marking result, includes:

[0037] Referring to the road segment attributes corresponding to each road segment included in the preprocessed road network data, the preprocessed road network data is compared with the historical preprocessed road network data with the segmentation results, and a first difference road segment, a second difference road segment, and a third difference road segment are determined in the historical preprocessed road network data with the segmentation results, wherein two road segment endpoints of the first difference road segment exist in both the historical preprocessed road network data with the segmentation results and the preprocessed road network data and the road segment attributes have not changed, two road segment endpoints of the second difference road segment exist in both the historical preprocessed road network data with the segmentation results and the preprocessed road network data and the road segment attributes have changed, and two road segment endpoints of the third difference road segment exist in the historical preprocessed road network data with the segmentation results but do not exist in the preprocessed road network data;

[0038] Performing a basic mark on the first difference road segment in the historical pre-processed road network data with the segmentation result, performing a change mark on the second difference road segment in the historical pre-processed road network data with the segmentation result, and performing a deletion mark on the third difference road segment in the historical pre-processed road network data with the segmentation result, wherein the basic mark is used to indicate that the road segment has not changed, the change mark is used to indicate that the road segment attribute has changed, and the deletion mark is used to indicate that the road segment does not exist;

[0039] Comparing the preprocessed road network data with the historical preprocessed road network data with the segmentation result, determining at least one fourth difference road segment in the preprocessed road network data, and adding a mark to the at least one fourth difference road segment in the preprocessed road network data, wherein two endpoints of the at least one fourth difference road segment do not exist in the historical preprocessed road network data with the segmentation result but exist in the preprocessed road network data, and the added mark is used to indicate that the road segment is a newly added road segment;

[0040] The marked pre-processed road network data and the marked historical pre-processed road network data with segmentation results are used as the difference marking results.

[0041] Optionally, performing partition processing on the pre-processed road network data according to the difference marking result to obtain the road network partition result includes:

[0042] According to the indication of the difference marking result, on the historical pre-processed road network data with the segmentation result, the partition result of the first difference road segment is kept unchanged, the segment attribute of the second difference road segment is updated, and the third difference road segment is deleted, thereby obtaining the updated historical pre-processed road network data with the segmentation result;

[0043] Referring to the updated historical pre-processed road network data with the segmentation results, partitioning the pre-processed road network data to obtain an initial partitioning result;

[0044] determining the at least one fourth difference road segment in the difference marking result, and dividing the at least one fourth difference road segment into at least one road segment group according to a connectivity relationship between the at least one fourth difference road segment;

[0045] The at least one road segment group is used to update the initial partitioning result, and the updated initial partitioning result is used as the road network partitioning result.

[0046] Optionally, the updating the initial partitioning result by using the at least one road segment group includes:

[0047] For each of the road section groups, query the historical partition results corresponding to a plurality of boundary nodes on the boundary of the road section group in the historical pre-processed road network data with the segmentation results;

[0048] When the historical partitioning result indicates that the plurality of boundary nodes are divided into the same first designated partition, in the initial partitioning result, all the road segments included in the road segment group are divided into the first designated partition;

[0049] When the historical partitioning result indicates that the plurality of boundary nodes are divided into at least two second designated partitions, the number of nodes of the boundary nodes divided into each of the second designated partitions is counted respectively; if the number of nodes of the at least two second designated partitions is different, in the initial partitioning result, all the road segments included in the road segment group are divided into the second designated partition with the highest number of nodes among the at least two second designated partitions; and if the number of nodes of the at least two second designated partitions is the same, in the initial partitioning result, all the road segments included in the road segment group are randomly divided into any second designated partition among the at least two second designated partitions;

[0050] When the historical partitioning result indicates that the plurality of boundary nodes are not divided into any partition, a new independent partition is constructed for the road segment group in the initial partitioning result.

[0051] According to a second aspect of the present application, a road network partitioning device is provided, the device comprising:

[0052] A preprocessing module is used to determine the road network data to be partitioned, and remove the idle road network data from the road network data to be partitioned to obtain preprocessed road network data;

[0053] a fusion module, configured to obtain instant delivery scenario data matching the road network data to be partitioned, and set a road segment attribute for each road segment included in the pre-processed road network data according to the instant delivery scenario data;

[0054] The partitioning module is used to partition the pre-processed road network data according to the road segment attributes corresponding to each road segment included in the pre-processed road network data to obtain a road network partitioning result.

[0055] Optionally, the preprocessing module is used to match the historical order data of the road network data to be partitioned within a specified historical time period, aggregate the delivery trajectory data in the historical order data, and determine at least one delivery activity area in the road network data to be partitioned; determine a preset expansion distance, expand the range of each delivery activity area according to the expansion distance, and use the at least one expanded delivery activity area as at least one commonly used road network area; use the road network data in the road network data to be partitioned that is located between the at least one commonly used road network area as the idle road network data, and eliminate the idle road network data, and use the road network data to be partitioned after elimination as the preprocessed road network data.

[0056] Optionally, the fusion module is used to obtain the instant delivery scene data that matches the road network data to be partitioned, and the instant delivery scene data includes but is not limited to business district range data, delivery range data, and point of interest area AOI boundary data; perform spatial intersection processing on the instant delivery scene data and the pre-processed road network data, determine at least one first road section that intersects with the instant delivery scene data in the road sections included in the pre-processed road network data, and use a preset intersection identifier to set the road section attribute for the at least one first road section; determine at least one in the pre-processed road network data. a second road section, and setting the road section attributes for the at least one second road section using a preset non-intersecting identifier, wherein the at least one second road section is a road section other than the at least one first road section in the road section included in the preprocessed road network data; according to the instant delivery scenario data, counting the road section flow data of each road section included in the preprocessed road network data, and determining the road section type data of each road section; based on the road section flow data and the road section type data of each road section, adjusting the road section attributes currently set for the road section included in the preprocessed road network data, and completing the setting of the road section attributes.

[0057] Optionally, the fusion module is used to determine at least one road section to be adjusted among the road sections included in the pre-processed road network data, and read the road section attributes currently set for each road section to be adjusted, wherein the road section flow data corresponding to the at least one road section to be adjusted is lower than a preset flow threshold and the corresponding road section type data belongs to a preset special road section type; determine the first road section to be adjusted among the at least one road section to be adjusted whose road section attributes are currently set to the intersection mark, and adjust the road section attributes of the first road section to be adjusted to the preset special mark; determine the second road section to be adjusted among the at least one road section to be adjusted whose road section attributes are currently set to the non-intersection mark, and adjust the road section attributes of the second road section to be adjusted to the intersection mark.

[0058] Optionally, the partitioning module is used to perform local road network merging on the pre-processed road network data to obtain pre-segmented road network data, and obtain a preset weight setting strategy, respectively query the weight value corresponding to the section attribute of each road section in the preset weight setting strategy, set the section weight for each road section according to the queried weight value, adopt a preset section segmentation algorithm, segment the pre-segmented road network data according to the section weight corresponding to each road section, obtain the pre-segmented road network data with the segmentation result, and restore the pre-segmented road network data with the segmentation result to the pre-processed road network data with the segmentation result. The preprocessed road network data with the segmentation result is used as the road network partitioning result; or, the historical road network partitioning result of the road network data to be partitioned is read, and the historical preprocessed road network data with the segmentation result with the smallest time interval between the segmentation time point and the current time point is extracted from the historical road network partitioning result, and the preprocessed road network data and the historical preprocessed road network data with the segmentation result are differentially marked with reference to the road segment attributes corresponding to each road segment included in the preprocessed road network data to obtain a differential marking result, and the preprocessed road network data is partitioned according to the differential marking result to obtain the road network partitioning result.

[0059] Optionally, the partitioning module is used to randomly select any road section node from the road section nodes included in the pre-processed road network data as the target road section node, perform breadth-first search with the target road section node as the starting point, and when a preset number of specified road section nodes are searched, use the road network composed of the preset number of specified road section nodes as a sub-road network; sort the preset number of specified road section nodes according to the order in which they are searched, and divide the sorted preset number of specified road section nodes into a source node group and a sink node group according to a preset division ratio; according to the source node group and the sink node group, use the maximum flow minimum cut algorithm to divide the sub-road network into two temporary road networks, and use the temporary road network including the target road section node in the two temporary road networks as a local road network, and use the road section connected to the local road network in the pre-processed road network data as the boundary section of the local road network; when the local road network meets the preset merging conditions, calculate the average value of the node coordinates of all road section nodes included in the local road network, and calculate the average value of the node coordinates obtained as temporary node coordinates, and creating a temporary road section node on the temporary node coordinates, wherein the preset merging conditions include but are not limited to the density of road network nodes and the parameter threshold of the maximum flow minimum cut algorithm; establishing a temporary road section between the temporary road section node and the boundary road section, and deleting all road section nodes and all road sections included in the local road network from the preprocessed road network data to complete the merging operation of the local road network; randomly selecting a road section node from other road section nodes included in the preprocessed road network data as a new target road section node, re-performing a breadth-first search with the new target road section node as the starting point and re-determining a new local road network, performing a merging operation on the new local road network until there is no selectable road section node in the preprocessed road network data, completing the operation of merging the local road network of the preprocessed road network data, and obtaining the pre-split road network data, wherein the other road section nodes are road section nodes in the preprocessed road network data other than the target road section node and the preset number of designated road section nodes.

[0060] Optionally, the partitioning module is used to determine at least one temporary road section node in the pre-segmented road network data with the segmentation results, and query the local road network corresponding to each temporary road section node in the pre-processed road network data, wherein the node coordinates of the temporary road section node are obtained by calculating the average value of the node coordinates of all road section nodes included in the corresponding local road network; the local road network corresponding to each temporary road section node is re-added to the pre-segmented road network data with the segmentation results, and each temporary road section node and the temporary road section established based on each temporary road section node are deleted, and the pre-segmented road network data with the segmentation results after the addition and deletion processing are used as the pre-processed road network data with the segmentation results.

[0061] Optionally, the partitioning module is used to refer to the segment attributes corresponding to each road segment included in the preprocessed road network data, compare the preprocessed road network data with the historical preprocessed road network data with the segmentation results, and determine the first difference segment, the second difference segment and the third difference segment in the historical preprocessed road network data with the segmentation results, wherein the two segment endpoints of the first difference segment exist in both the historical preprocessed road network data with the segmentation results and the preprocessed road network data and the segment attributes have not changed, the two segment endpoints of the second difference segment exist in both the historical preprocessed road network data with the segmentation results and the preprocessed road network data and the segment attributes have changed, and the two segment endpoints of the third difference segment exist in the historical preprocessed road network data with the segmentation results but do not exist in the preprocessed road network data; perform basic marking on the first difference segment in the historical preprocessed road network data with the segmentation results, and The second difference road segment is marked as changed in the preprocessed road network data, and the third difference road segment is marked as deleted in the historical preprocessed road network data with the segmentation result, wherein the basic mark is used to indicate that the road segment has not changed, the change mark is used to indicate that the road segment attribute has changed, and the deletion mark is used to indicate that the road segment does not exist; the preprocessed road network data is compared with the historical preprocessed road network data with the segmentation result, at least one fourth difference road segment is determined in the preprocessed road network data, and the at least one fourth difference road segment is marked as added in the preprocessed road network data, wherein the two endpoints of the at least one fourth difference road segment do not exist in the historical preprocessed road network data with the segmentation result but exist in the preprocessed road network data, and the addition mark is used to indicate that the road segment is a newly added road segment; the preprocessed road network data with the completed marking and the historical preprocessed road network data with the segmentation result with the completed marking are used as the difference marking results.

[0062] Optionally, the partitioning module is used to keep the partitioning result of the first difference road section unchanged on the historical preprocessed road network data with the segmentation result in accordance with the indication of the difference marking result, update the segment attributes of the second difference road section, and delete the third difference road section to obtain the updated historical preprocessed road network data with the segmentation result; partition the preprocessed road network data with reference to the updated historical preprocessed road network data with the segmentation result to obtain an initial partitioning result; determine the at least one fourth difference road section in the difference marking result, and divide the at least one fourth difference road section into at least one segment group according to the connectivity relationship between the at least one fourth difference road sections; use the at least one segment group to update the initial partitioning result, and use the updated initial partitioning result as the road network partitioning result.

[0063] Optionally, the partitioning module is used to query the historical partitioning results corresponding to the multiple boundary nodes on the boundary of the road section group in the historical pre-processed road network data with the segmentation results for each of the road section groups; when the historical partitioning results indicate that the multiple boundary nodes are divided into the same first specified partition, in the initial partitioning results, all the road sections included in the road section group are divided into the first specified partition; when the historical partitioning results indicate that the multiple boundary nodes are divided into at least two second specified partitions, the number of nodes of the boundary nodes divided into each of the second specified partitions is counted respectively, and the number of nodes of the boundary nodes divided into the at least two second specified partitions is counted respectively. When the number of nodes of the second designated partitions is different, in the initial partitioning result, all the road sections included in the road section group are divided into the second designated partition with the highest number of nodes among the at least two second designated partitions; and when the number of nodes of the at least two second designated partitions is the same, in the initial partitioning result, all the road sections included in the road section group are randomly divided into any second designated partition among the at least two second designated partitions; when the historical partitioning result indicates that the multiple boundary nodes are not divided into any partition, in the initial partitioning result, a new independent partition is constructed for the road section group.

[0064] According to a third aspect of the present application, a computer device is provided, comprising a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of any one of the methods described in the first aspect when executing the computer program.

[0065] According to a fourth aspect of the present application, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps of any one of the methods in the first aspect are implemented.

[0066] By means of the above technical solution, the present application provides a road network partitioning method, device, computer equipment and computer-readable storage medium. The present application determines the road network data to be partitioned, eliminates the idle road network data in the road network data to be partitioned, obtains pre-processed road network data, obtains instant delivery scenario data matching the road network data to be partitioned, sets section attributes for each road section included in the pre-processed road network data according to the instant delivery scenario data, partitions the pre-processed road network data according to the section attributes corresponding to each road section included in the pre-processed road network data, and obtains a road network partitioning result. By integrating the instant delivery scenario data, both natural cuts and the characteristics of the instant delivery scenario are taken into account in the road network partitioning process. At the same time, by eliminating idle road network data, the scale of the road network to be partitioned is reduced, and the efficiency of road network partitioning is improved. The rationality and flexibility of road network partitioning are good, thereby ensuring the accuracy and efficiency of subsequent path planning.

[0067] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0068] Various other advantages and benefits will become apparent to those skilled in the art upon reading the detailed description of the preferred embodiment below. The accompanying drawings are for illustration purposes only and are not to be considered as limiting the present application. The same reference symbols are used throughout the drawings to represent the same components. In the drawings:

[0069] Figure 1 A schematic diagram of a road network partitioning method provided in an embodiment of the present application is shown;

[0070] Figure 2A A schematic diagram of another road network partitioning method provided in an embodiment of the present application is shown;

[0071] Figure 2B A schematic diagram of a road network partitioning method provided in an embodiment of the present application is shown;

[0072] Figure 2C A schematic diagram of a process flow of another road network partitioning method provided in an embodiment of the present application is shown;

[0073] Figure 2D A schematic diagram of a process flow of another road network partitioning method provided in an embodiment of the present application is shown;

[0074] Figure 3 A schematic structural diagram of a road network partitioning device provided in an embodiment of the present application is shown;

[0075] Figure 4 A schematic diagram of the device structure of a computer device provided in an embodiment of the present application is shown. DETAILED DESCRIPTION

[0076] The following describes exemplary embodiments of the present application in more detail with reference to the accompanying drawings. Although exemplary embodiments of the present application are shown in the accompanying drawings, it should be understood that the present application can be implemented in various forms and should not be limited by the embodiments set forth herein. Rather, these embodiments are provided to enable a more thorough understanding of the present application and to fully convey the scope of the present application to those skilled in the art.

[0077] The present application embodiment provides a road network partitioning method, such as Figure 1 As shown, the method includes:

[0078] 101. Determine the road network data to be partitioned, remove the idle road network data from the road network data to be partitioned, and obtain pre-processed road network data.

[0079] In the embodiment of the present application, it is necessary to first determine the road network data to be processed, and use this road network data as the road network data to be partitioned. The road network data to be partitioned includes a relatively large-scale mixed cycling and walking road network.

[0080] The applicant recognizes that in an instant delivery scenario, not all road networks are frequently used. When partitioning a road network in an instant delivery scenario, if the entire road network is included in the partitioning logic, it will be a significant challenge in terms of both performance and storage space. Since the service scope of an instant delivery scenario tends to be local (roughly within a few kilometers), it is generally concentrated in urban centers with a relatively high population density. Furthermore, due to limitations such as the delivery range, cross-regional orders are rarely seen. Therefore, to reduce processing difficulty and resource consumption, the embodiments of this application will first pre-process the partitioned road network data, primarily removing idle road network data that will rarely be used.

[0081] Idle road network data refers to road sections that are rarely used in instant delivery services. These sections are often located in remote areas or on the outskirts of cities, where delivery demand is low. By identifying and removing this idle road network data, more streamlined pre-processed road network data can be obtained, significantly reducing the amount of data required for subsequent processing and improving overall processing efficiency. For example, in some remote towns on the outskirts of cities, their roads are rarely used in instant delivery and can therefore be removed as idle road networks.

[0082] 102. Obtain instant delivery scenario data that matches the road network data to be partitioned, and set the segment attributes for each road segment included in the pre-processed road network data according to the instant delivery scenario data.

[0083] To make the road network data to be partitioned more suitable for instant delivery scenarios, the present embodiment further integrates instant delivery scenario data based on the pre-processed road network data. The instant delivery scenario data includes but is not limited to business district range data, delivery range data, AOI (area of interest) boundary data, etc.

[0084] When integrating instant delivery scenario data, spatial intersection processing is used to determine whether each road segment in the preprocessed road network data intersects with the instant delivery scenario data. Intersecting road segments indicate their importance in instant delivery and require segment attribute settings. Non-intersecting road segments also have their segment attributes set to reflect their minimal impact on the instant delivery scenario in subsequent partitioning operations, providing an important basis for subsequent partitioning.

[0085] 103. Based on the road segment attributes corresponding to each road segment included in the pre-processed road network data, the pre-processed road network data is partitioned to obtain a road network partition result.

[0086] After completing the preprocessing and scene data fusion, the embodiment of the present application will partition the preprocessed road network data according to the road section attributes to obtain the road network partition results.

[0087] Among them, in order to achieve efficient and reasonable partitioning, the embodiment of the present application adopts full segmentation or incremental segmentation and a combination of the two to perform partitioning. Full segmentation can obtain a relatively reasonable segmentation result based on the principle of overall optimization and local adjustment as needed. However, full segmentation takes a long time. For frequently changing road network data, performing full segmentation for each change will cause a waste of resources. Therefore, on the basis of full segmentation, the embodiment of the present application can also perform incremental segmentation on the different road networks by identifying the differences between the two versions of the road network. This segmentation method is fast and can avoid unnecessary waste of resources. In the process of actual application, full segmentation or incremental segmentation can be selected according to the specific scenario, and the embodiment of the present application does not specifically limit this.

[0088] For example, in the case of significant changes in the road network, such as large-scale urban construction or traffic planning adjustments, full segmentation can be selected; in the case of minor changes in the road network, such as local road maintenance or the addition of a small number of roads, incremental segmentation can be selected. In addition, to maintain the timeliness and accuracy of the road network data, the embodiment of the present application will perform a full segmentation on a regular basis (such as monthly or weekly), and then perform incremental segmentation based on demand. This will improve the partitioning efficiency of the road network data while ensuring more accurate road network support for instant delivery services.

[0089] The method provided in the embodiment of the present application determines the road network data to be partitioned, eliminates the idle road network data in the road network data to be partitioned, obtains pre-processed road network data, obtains instant delivery scenario data matching the road network data to be partitioned, sets section attributes for each road section included in the pre-processed road network data according to the instant delivery scenario data, partitions the pre-processed road network data according to the section attributes corresponding to each road section included in the pre-processed road network data, and obtains a road network partitioning result. By integrating the instant delivery scenario data, both natural cuts and the characteristics of the instant delivery scenario are taken into account in the road network partitioning process. At the same time, by eliminating idle road network data, the scale of the road network to be partitioned is reduced, and the efficiency of road network partitioning is improved. The rationality and flexibility of road network partitioning are good, thereby ensuring the accuracy and efficiency of subsequent path planning.

[0090] Furthermore, as a refinement and extension of the specific implementation of the above embodiment, in order to fully illustrate the specific implementation process of this embodiment, the embodiment of the present application provides another road network partitioning method, such as Figure 2A As shown, the method includes:

[0091] 201. Determine the road network data to be partitioned, remove the idle road network data from the road network data to be partitioned, and obtain pre-processed road network data.

[0092] In the context of instant delivery, high-quality route planning services are the foundation of this service. This is a complex, comprehensive metric: the more accurate the route planning, the closer it is to real-world instant delivery. Real-time performance is a crucial metric in route planning services, especially in this context. Real-time visibility into road conditions significantly impacts merchants, users, and drivers.

[0093] Currently, high-performance path planning algorithms are the mainstream in most scenarios, and road network partitioning, based on the divide-and-conquer principle, is a key support for this. Traditional road network partitioning primarily identifies geographical features such as rivers and mountains. However, instant delivery scenarios are not only sensitive to natural geographical features, but also to data unique to instant delivery scenarios, such as business districts, delivery ranges, area of interest (AOI), and road network attributes. Traditional road network partitioning cannot utilize these features, which affects the experience of path planning services in instant delivery scenarios. In addition, due to the complexity of the road network, the performance of traditional road network partitioning is even more challenging when applied to local life scenarios.

[0094] Therefore, the embodiment of the present application proposes a road network partitioning method based on the characteristics of the instant delivery scenario. On the one hand, it follows the traditional road network partitioning method and takes into account the natural geographical characteristics. On the other hand, it combines the characteristics of the instant delivery scenario such as business district, delivery range, AOI, road network attributes, etc., and comprehensively considers these two types of information to make the road network partitioning result more in line with the instant delivery scenario; and, based on the instant delivery scenario, some idle road network data is eliminated, the road network scale is reduced, the performance bottleneck of road network partitioning is alleviated, and the flexibility of road network partitioning is improved.

[0095] In the preliminary stage of road network partitioning, the embodiment of the present application will first determine the road network data to be partitioned, which includes basic road network information such as road name, location coordinates, road grade, and number of lanes; and in order to achieve the integration of instant delivery scenario data and road network data, the embodiment of the present application will further match the historical order data of the road network data to be partitioned within a specified historical time period (such as the past six months), and use the historical order data to optimize the road network data to be partitioned.

[0096] After matching the historical order data of the road network data to be partitioned within the specified historical time period, rich delivery trajectory data can be obtained in the historical order data. The embodiment of the present application will use an aggregation algorithm to aggregate the delivery trajectory data, and determine at least one delivery activity area in the road network data to be partitioned to reveal the hot spots of rider activities, that is, the road network range mainly involved in the historical orders. However, in order to ensure the accuracy and practicality of the road network partitioning, the embodiment of the present application does not just stop at these hot spots, but further considers their peripheral areas. By presetting an expansion distance (such as 5 kilometers), the embodiment of the present application expands the range of each delivery activity area according to the expansion distance, appropriately expands the range of each delivery activity area, and uses at least one expanded delivery activity area as at least one commonly used road network area to avoid the impact at the range boundary, so as to ensure that the commonly used road network area covers possible boundary effects and potential areas of rider activities.

[0097] Considering that if a certain section of road network has not been used by riders for a long time, it may not be necessary in future delivery activities. Therefore, after obtaining the commonly used road network area, the embodiment of the present application will treat the road network data located between at least one commonly used road network area in the road network data to be partitioned as idle road network data, and eliminate the idle road network data, and use the eliminated road network data to be partitioned as pre-processed road network data, so that the idle road networks that will not be used for a long time between the commonly used road networks are eliminated, and pre-processed road network data containing only commonly used road networks is obtained. By eliminating these idle road network data, a more streamlined and efficient pre-processed road network data can be obtained, which can greatly reduce the amount of data to be processed later and improve computing efficiency while ensuring that the road network partitioning is more in line with actual delivery needs.

[0098] Specifically, see Figure 2B , where the map includes three circle areas. The inner circle of each circle area is the distribution activity area determined based on historical order data, and the outer circle of each circle area is the commonly used road network area obtained by expanding the distribution activity area. In this way, the pre-processed road network data can be obtained by removing the idle road network data between these three commonly used road network areas on the map. It should be noted that in the actual application process, whether it is the road network range obtained after aggregation or the road network range of the commonly used road network obtained in the end, it is actually generally an irregular polygon. Figure 2B In the example, the range is simplified to a circle for ease of description. This process can achieve the goal of initially reducing the size of the road network without affecting actual road network planning services. In actual application, this process can reduce the size of the national road network by over 30%, resulting in preprocessed road network data.

[0099] 202. Obtain instant delivery scenario data that matches the road network data to be partitioned, and set segment attributes for each road segment included in the pre-processed road network data according to the instant delivery scenario data.

[0100] In order to make the road network segmentation results more in line with the instant delivery scenario, before entering the road network segmentation, the implementation of this application will first integrate the instant delivery scenario data with the characteristics of the instant delivery scenario, and give each road section in the road network an attribute label that is more in line with the delivery needs, thereby providing more accurate data support for subsequent road network segmentation.

[0101] First, in the embodiments of this application, it is necessary to obtain real-time delivery scenario data that matches the road network data to be partitioned. Real-time delivery scenario data includes but is not limited to business district range data, delivery range data, and AOI boundary data. AOI, or Area of Interest, is divided according to actual divisions such as commercial districts and residential areas. For example, a residential complex or a shopping mall can be called an AOI. Roads within the AOI are called AOI internal roads, and roads outside the AOI are called AOI external roads.

[0102] Subsequently, it is necessary to perform spatial intersection processing on the instant delivery scenario data and the pre-processed road network data, determine at least one first road segment that intersects with the instant delivery scenario data among the road segments included in the pre-processed road network data, and use a preset intersection identifier to set the road segment attributes for the at least one first road segment. This process aims to find road segments that are closely related to instant delivery activities, that is, those road segments that intersect with data such as business districts, distribution, and range AOI boundaries. Furthermore, it is also necessary to determine at least one second road segment in the pre-processed road network data, and use a preset non-intersection identifier to set the road segment attributes for the at least one second road segment. The at least one second road segment is a road segment other than the at least one first road segment included in the pre-processed road network data, that is, to find road segments that are not related to instant delivery activities.

[0103] Next, we set specific intersection and non-intersection flags for these road segments to distinguish the first and second road segments. In practice, we can set the intersection flag to 1 and the non-intersection flag to 0. That is, in the pre-processed road network data, the road segments that intersect with the instant delivery scenario data are marked as 1, and the remaining non-intersecting road segments are marked as 0.

[0104] On the basis of the above process, the embodiment of the present application will further adjust the road section attributes based on the road section flow data and the road attribute data of each road section. The road section flow data is obtained by analyzing the rider trajectory data over the past period of time, which reflects the actual use of each road section in distribution activities. Road attribute data is the basic characteristics of each road, such as whether it is rideable, whether it is an AOI internal road, whether it is a bridge, road grade, etc., where the road grade is used to measure the status of the road in the road network, such as expressways, urban expressways, urban main roads, urban secondary roads, etc.; combined with road attribute data, the road section attributes can be set more finely, such as assigning different attribute values to road sections based on indicators such as traffic density, congestion level, and average speed. Specifically, it is necessary to first collect the traffic data of each road section included in the pre-processed road network data and determine the road type data of each road section based on the instant delivery scenario data; then, based on the traffic data and road type data of each road section, adjust the road attributes currently set for the road section included in the pre-processed road network data to complete the setting of the road attribute. The process of adjusting the road attribute is as follows:

[0105] First, at least one road segment to be adjusted is identified among the road segments included in the preprocessed road network data, and the currently set road segment attributes for each road segment to be adjusted are read. The road segment flow data corresponding to the at least one road segment to be adjusted is below a preset flow threshold, and the corresponding road segment type data belongs to a preset special road segment type. Subsequently, a first road segment to be adjusted is identified among the at least one road segment to be adjusted, whose road segment attributes are currently set to an intersection indicator, and the road segment attributes of the first road segment to be adjusted are adjusted to the preset special indicator. A second road segment to be adjusted is identified among the at least one road segment to be adjusted, whose road segment attributes are currently set to a non-intersection indicator, and the road segment attributes of the second road segment to be adjusted are adjusted to the intersection indicator. In other words, the road segments included in the preprocessed road network data are first identified for which road segment flow data is below a preset flow threshold. The road segment types corresponding to these road segments are then checked. If the road segments to be adjusted have low flow rates, and their road segment types indicate a relatively high road grade, and are special road segments such as bridges, tunnels, or ferries, these road segments are identified as road segments to be adjusted and continue to be identified as having the currently set attribute values for the road segments to be adjusted. Among them, if the original section attribute is 1 (i.e., intersection mark), then since the level of the section to be adjusted is higher and needs more attention, the section attribute of the section to be adjusted will be re-marked as 2, which is used to indicate a special mark; if the original section attribute is 0, then since the level of the section to be adjusted is higher and needs more attention, the section attribute of the section to be adjusted will be re-marked as 1. The above content is the adjustment process for sections whose section traffic data is lower than the preset traffic threshold and the corresponding section type data belongs to the preset special section type. In actual application, if a section has high traffic, low road grade, and is not a special type of section such as a bridge, tunnel, or ferry, then its original section attribute settings can be maintained.

[0106] In this way, by introducing instant delivery scenario data, the results of subsequent road network partitioning can not only conform to natural characteristics, but also be more in line with the characteristics of instant delivery scenarios, which helps to improve delivery efficiency and service quality.

[0107] 203. Perform local road network merging on the pre-processed road network data to obtain pre-segmented road network data.

[0108] After preprocessing the road network data, even though a series of optimization processes have been performed on the original road network data, the scale of the preprocessed road network may still be quite large. In order to further reduce the scale of the road network while ensuring that the results of the road network segmentation are not affected, the embodiment of the present application draws on some of the ideas of the PUNCH algorithm and performs a local road network merging operation on the preprocessed road network. The PUNCH algorithm is a road network segmentation algorithm that first identifies small natural cuts and then merges these cuts to obtain the final road network segmentation result. The specific merging process is as follows:

[0109] First, a random segment node from the preprocessed network data is selected as the target segment node. A breadth-first search is performed starting from the target segment node. Once a preset number of specified segment nodes are found, the network consisting of these nodes is considered a subnetwork. Breadth-first search is a classic graph search algorithm that begins at an initial node and searches the graph layer by layer until the target node or a certain number of nodes are found. For example, a segment node v is randomly selected from the preprocessed network data as the center point. A breadth-first search is performed starting from v, stopping when u specified segment nodes are traversed, resulting in a subnetwork g.

[0110] Next, the preset number of designated road segment nodes are sorted in the order they were found during the search, and the sorted preset number of designated road segment nodes are divided into source node groups and sink node groups according to a preset division ratio. Continuing with the above example, the u designated road segment nodes are sorted in the order they were found, with a certain proportion of points at the front being used as sources and a certain proportion of points at the back being used as sinks.

[0111] Then, based on the source node group and sink node group, the maximum flow minimum cut algorithm is used to partition the sub-network into two temporary networks. The temporary network that includes the target segment node in the two temporary networks is considered the local network, and the segments connected to the local network in the pre-processed network data are considered the boundary segments of the local network. The maximum flow minimum cut algorithm is a classic algorithm in the field of graph segmentation. It divides the entire graph into two relatively independent sub-graphs by calculating the minimum cut between the source and the sink in the graph. After the network is partitioned, the segments connecting each sub-network are called cuts. Continuing with the above example, the maximum flow minimum cut algorithm is used to partition the sub-graph into two temporary networks. The temporary network containing the segment node v is called the local network, and the segments connected to the local network in the pre-processed network data are called the boundary segments of the local network.

[0112] Finally, the merging of the local road networks begins. Specifically, in the embodiment of the present application, preset merging conditions are set, which include but are not limited to the density of the road network nodes, the parameter threshold of the maximum flow minimum cut algorithm, etc. When the local road network meets the preset merging conditions, the embodiment of the present application will calculate the average value of the node coordinates of all the segment nodes included in the local road network, use the calculated average coordinates as temporary node coordinates, and create a temporary segment node on the temporary node coordinates; at the same time, establish a temporary segment between the temporary segment node and the boundary segment, and delete all the segment nodes and all the segments included in the local road network from the pre-processed road network data, completing the merging operation of the local road network. Continuing with the above example, if the local road network meets the preset merging conditions, all the segment nodes and all the segments in the local road network are deleted from the pre-processed road network data and merged into a temporary node v'. The temporary node coordinates of the temporary node v' are obtained by calculating the average value of the coordinates of the points in the local road network, and a temporary segment is established between the temporary node v' and the boundary segment of the local road network, completing the merging of the local road network.

[0113] After completing a merge through the above process, a new target road segment node can be randomly selected from the other road segment nodes included in the pre-processed road network data. A breadth-first search is then performed again starting from the new target road segment node to redefine a new local road network. The merge operation is then performed on the new local road network until no selectable road segment nodes exist in the pre-processed road network data. This completes the local road network merge operation on the pre-processed road network data and yields pre-segmented road network data. The "other road segment nodes" are road segment nodes in the pre-processed road network data other than the target road segment node and a preset number of designated road segment nodes. In other words, each time the pre-processed road network data is iteratively merged, nodes in the local road network that have already been merged no longer need to be processed as central nodes and can be directly skipped. Therefore, it is not actually necessary to traverse all nodes in the entire pre-processed road network.

[0114] In this way, by merging local road networks, the road network scale is further reduced on the basis of pre-processing road network data, and the efficiency of road network segmentation is improved; according to the above process, the national road network scale can be further reduced by more than 50% on the basis of road network pre-processing, and pre-segmented road network data can be obtained.

[0115] 204. Obtain the preset weight setting strategy, query the weight value corresponding to the section attribute of each road section in the preset weight setting strategy respectively, set the section weight for each road section according to the queried weight value, use the preset section segmentation algorithm, segment the pre-segmented road network data according to the section weight corresponding to each road section, and obtain the pre-segmented road network data with the segmentation results.

[0116] After further reducing the scale of the road network, the embodiment of the present application will make full use of the characteristics of the pre-cut road network data, combined with the road network cutting algorithm, to achieve a fine division of the road network. When cutting, in order to reflect the importance of each road section, traffic congestion, distribution demand distribution, etc., it is necessary to obtain a preset weight setting strategy, query the weight value corresponding to the road section attribute of each road section in the preset weight setting strategy, and set the road section weight for each road section according to the queried weight value. Among them, when setting the road section weight, the larger the value of the road section attribute, the smaller the road section weight assigned, and vice versa. The weight between the road sections can be set to an integer, for example, the minimum is set to 1 and the maximum is set to 10. According to this step, the weight of each road section is obtained, and the road network cutting used in the subsequent steps is set to a relatively small weight, which will make these road sections tend to be cut when the road network is cut, that is, these road sections with smaller weights are more likely to be cut.

[0117] After completing the setting of the road section weights, the preset road section segmentation algorithm can be used to segment the pre-segmented road network data according to the road section weights corresponding to each road section, and obtain the pre-segmented road network data with segmentation results. The preset road section segmentation algorithm can be the Inertial Flow algorithm, which is an efficient road network segmentation algorithm that uses embedding technology and sorts nodes according to certain rules, and then obtains the road network segmentation results through the maximum flow minimum cut algorithm. The Inertial Flow algorithm has fewer parameters and uniform segmentation results, which is suitable for processing pre-segmented road network data. Specifically, the InertialFlow algorithm will perform multi-level segmentation on the entire graph through recursive bisection, and finally obtain a reasonable road network partitioning result.

[0118] This process allows for a finer division of the pre-segmented road network data, ensuring that each partition covers key delivery areas while maintaining reasonable traffic flow and delivery efficiency. By assigning different weights to different road sections during the segmentation process, the resulting road network segmentation better aligns with actual delivery needs, helping to improve delivery efficiency and service quality.

[0119] 205. Restore the pre-segmented road network data with the segmentation results to pre-processed road network data with the segmentation results, and use the pre-processed road network data with the segmentation results as the road network partitioning results.

[0120] The above-mentioned road network merging process is actually intended to reduce the scale of the road network and make the road network partitioning more efficient. In actual application, the partitioned road network is needed for subsequent services such as route planning, and these services still rely on detailed road network information. Therefore, to ensure the integrity of the road network partitioning results, in the embodiment of the present application, the pre-segmented road network data with the segmentation results will be restored to pre-processed road network data with the segmentation results, and the pre-processed road network data with the segmentation results will be used as the road network partitioning results, ensuring that the road network partitioning results include the original detailed road network information, providing strong data support for subsequent road network planning.

[0121] In the specific implementation process, it is necessary to traverse each road section node in the pre-segmented road network data with the segmentation results, and determine at least one temporary road section node in the pre-segmented road network data with the segmentation results. Then, the local road network corresponding to each temporary road section node is queried in the pre-processed road network data, wherein the node coordinates of the temporary road section node are obtained by calculating the average value of the node coordinates of all road section nodes included in the corresponding local road network. In other words, the temporary road section node is the temporary road section node constructed in the above-mentioned step 203. Subsequently, the local road network corresponding to each temporary road section node is re-added to the pre-segmented road network data with the segmentation results, and each temporary road section node and the temporary road section established based on each temporary road section node are deleted. The pre-segmented road network data with the segmentation results after the addition and deletion process is used as the pre-processed road network data with the segmentation results.

[0122] That is to say, in the process of traversing the road segment nodes, if a road segment node belongs to a temporary road segment node, the points and road segments in the original local road network corresponding to the temporary road segment node are re-added to the pre-divided road network, and the temporary road segment node and the temporary road segment between the temporary road segment node and the boundary section of the original local road network are deleted; if a road segment node does not belong to a temporary road segment node, it is skipped and no processing is performed.

[0123] The method provided in the embodiment of the present application determines the road network data to be partitioned, eliminates the idle road network data in the road network data to be partitioned, obtains pre-processed road network data, obtains instant delivery scenario data matching the road network data to be partitioned, sets section attributes for each road section included in the pre-processed road network data according to the instant delivery scenario data, partitions the pre-processed road network data according to the section attributes corresponding to each road section included in the pre-processed road network data, and obtains a road network partitioning result. By integrating the instant delivery scenario data, both natural cuts and the characteristics of the instant delivery scenario are taken into account in the road network partitioning process. At the same time, by eliminating idle road network data, the scale of the road network to be partitioned is reduced, and the efficiency of road network partitioning is improved. The rationality and flexibility of road network partitioning are good, thereby ensuring the accuracy and efficiency of subsequent path planning.

[0124] The above embodiment obtains the final pre-processed road network data with segmentation results through road network pre-segmentation, road network segmentation, and road network segmentation results. These steps involve the entire road network being input for segmentation, which is a full segmentation method. Full segmentation produces a more reasonable segmentation result based on overall optimization and local adjustments as needed. However, this segmentation method is time-consuming, and if only a small amount of road network changes are involved, re-performing a full segmentation results in a waste of resources. Therefore, after obtaining the results of a full segmentation, incremental segmentation can be performed based on this. Incremental segmentation is a segmentation method that identifies the differences between two versions of the road network based on a full segmentation and segments the different road networks. Incremental segmentation is fast and can avoid the unnecessary waste of resources caused by frequent full segmentations. During actual segmentation, the choice of full or incremental segmentation can be based on the scenario. Generally, a full segmentation can be performed at regular intervals (e.g., monthly or weekly), and then incremental segmentation can be performed based on demand (daily, hourly, or even minutely). In addition, since the actual speed of road network changes is relatively fast, it is generally not recommended to not perform full segmentation for a long period of time (several months or even more than half a year). In order to fully illustrate the specific implementation process of incremental segmentation, the embodiment of this application provides another road network partitioning method, such as Figure 2C As shown, the method includes:

[0125] 206. Read the historical road network partitioning results of the road network data to be partitioned, and extract the historical pre-processed road network data with the partitioning results that has the smallest time interval between the partitioning time point and the current time point from the historical road network partitioning results.

[0126] In an embodiment of the present application, it is necessary to read the historical road network partitioning results of the road network data to be partitioned, and extract the historical pre-processed road network data with the partitioning results with the smallest time interval between the partitioning time point and the current time point from the historical road network partitioning results, that is, from the historical road network partitioning results, filter out the historical pre-processed road network data with the partitioning results closest to the current time point.

[0127] Among them, the historical road network partitioning results refer to the partitioning processing records of the road network at past time nodes. The segmentation time point refers to the time node in the historical road network partitioning results at which the road network data is actually segmented. Specifically, all historical road network partitioning results of the road network data to be partitioned can be read, and the historical road network partitioning results contain snapshots of road network partitions at multiple time points. Then, the time interval between the segmentation time point and the current time point of each historical road network partitioning result is calculated, and sorted according to the time interval. Finally, the historical road network partitioning result with the smallest time interval, that is, the historical pre-processed road network data with the segmentation results, is selected as a reference benchmark for subsequent processing. In this way, by utilizing the historical pre-processed road network data with the smallest time interval, a more reliable basis is provided for subsequent difference marking and partitioning processing.

[0128] 207. Referring to the road segment attributes corresponding to each road segment included in the pre-processed road network data, perform difference marking on the pre-processed road network data and the historical pre-processed road network data with the segmentation results to obtain a difference marking result.

[0129] In an embodiment of the present application, in order to accurately determine the changes that have occurred in the road network, it is necessary to refer to the road segment attributes corresponding to each road segment included in the pre-processed road network data, and perform difference marking on the pre-processed road network data and the historical pre-processed road network data with the segmentation results to obtain a difference marking result. The difference marking result also indicates the changes that have occurred in the road network. In actual application, the historical pre-processed road network data with the segmentation results can be used as the base version of the road network, and the pre-processed road network data can be used as the new version of the road network. On the basis of the base version of the road network, the changes in the new version of the road network are marked to obtain the diff between the two versions. Diff is a diff marking of road segments to identify differences such as addition, deletion, and attribute changes of road segments. The diff marking mainly includes four marking types: base, addition, deletion, and modification. The definitions of these four types are described below:

[0130] 1. Foundation.

[0131] "Base" indicates that the road segment exists in both the base and updated versions of the network. This means that both endpoints of the segment exist in both versions, and the road attributes of the segment remain unchanged. This means that the segment remains consistent in both versions and no modifications are required.

[0132] 2. Increase.

[0133] "Add" indicates that the road segment does not exist in the base version of the network but does in the updated version. This means that the two endpoints of the road segment may not exist in the base version of the network but do in the updated version. This means that network expansion or optimization requires adding a mark to the road segment in the updated version.

[0134] 3. Delete.

[0135] Deleted means that the road segment exists in the base version of the network but not in the updated version. This means that the two endpoints of the road segment exist in the base version of the network but not necessarily in the updated version. This may be due to road reconstruction, abandonment, or other reasons. In the base version of the network, the road segment will be marked as deleted.

[0136] 4. Change.

[0137] Changed means that the road segment exists in both the base version and the updated version. This means that both endpoints of the segment exist in both the base version and the updated version, but the segment's road attributes have changed. This means that in the updated version, the segment will be marked as changed and its attributes will be updated.

[0138] Based on the above definition, we can start to mark the differences between the processed road network data and the historical pre-processed road network data with the segmentation results. Specifically, we need to refer to the section attributes corresponding to each road section included in the pre-processed road network data, compare the pre-processed road network data with the historical pre-processed road network data with the segmentation results, and determine the first difference section, the second difference section and the third difference section in the historical pre-processed road network data with the segmentation results. Among them, the two segment endpoints of the first difference segment exist in both the historical preprocessed road network data with the segmentation results and the preprocessed road network data, and the segment attributes have not changed, that is, the first difference segment is the segment mentioned above that needs to be marked as "base"; the two segment endpoints of the second difference segment exist in both the historical preprocessed road network data with the segmentation results and the preprocessed road network data, and the segment attributes have changed, that is, the second difference segment is the segment mentioned above that needs to be marked as "modified"; the two segment endpoints of the third difference segment exist in the historical preprocessed road network data with the segmentation results and do not exist in the preprocessed road network data, that is, the third difference segment is the segment mentioned above that needs to be marked as "deleted".

[0139] After determining the above-mentioned difference road sections, the first difference road section can be marked as basic (i.e., base) in the historical pre-processed road network data with the segmentation results, the second difference road section can be marked as changed (i.e., changed) in the historical pre-processed road network data with the segmentation results, and the third difference road section can be marked as deleted (i.e., deleted) in the historical pre-processed road network data with the segmentation results, wherein the basic mark is used to indicate that the road section has not changed, the change mark is used to indicate that the road section attributes of the road section have changed, and the deletion mark is used to indicate that the road section does not exist.

[0140] However, the "add" flag has not yet been confirmed in the above process. Therefore, it is necessary to continue comparing the preprocessed road network data with the historical preprocessed road network data containing the segmentation results, determine at least one fourth difference road segment in the preprocessed road network data, and mark the at least one fourth difference road segment as "add" in the preprocessed road network data. The at least one fourth difference road segment has two endpoints that do not exist in the historical preprocessed road network data containing the segmentation results but exist in the preprocessed road network data. The "add" flag is used to indicate that the road segment is a newly added road segment. In other words, the fourth difference road segment is the road segment that needs to be marked as "add" as mentioned above.

[0141] After completing the above process, the marked pre-processed road network data and the marked historical pre-processed road network data with segmentation results can be used as the difference marking results. In this way, by identifying and marking the road network data diff, the differences between the two versions of the road network can be obtained, preparing for incremental segmentation.

[0142] 208. According to the difference marking results, the pre-processed road network data is partitioned to obtain a road network partition result.

[0143] The difference marking results record the differences between the new version of the road network and the base version, including the addition, deletion, and attribute changes of road sections. These differences are precisely categorized into four types: base, addition, deletion, and modification, providing clear guidance for subsequent zoning processing.

[0144] When partitioning the preprocessed road network data according to the difference marking results to obtain the network partition results, there is essentially no need to process the base, deletion, and modification types, as the partition results for these three types are already known (and can be directly obtained from the base version of the road network). The key is to pay attention to the topological changes after the deletion operation. Therefore, according to the instructions of the difference marking results, the partition results of the first difference segment (i.e., the base type segment) can be maintained unchanged in the historical preprocessed road network data with the partition results, reflecting the preservation of stable segments. The segment attributes of the second difference segment (i.e., the modified type segment) are updated to ensure the accuracy of the information in the new version of the road network. The third difference segment (i.e., the deletion type segment) is deleted, and special attention is paid to handling the resulting topological changes to ensure the integrity and connectivity of the road network structure. This results in the updated historical preprocessed road network data with the partition results. The preprocessed road network data is then partitioned with reference to the updated historical preprocessed road network data with the partition results to obtain the initial partition results.

[0145] For newly added data, it is first necessary to determine at least one fourth difference road segment in the difference marking result, and divide the at least one fourth difference road segment into at least one road segment group according to the connectivity relationship between the at least one fourth difference road segment. That is, it is necessary to group according to the connectivity relationship to ensure that the boundary points of each group are the boundary points of the original graph or the nodes with existing segmentation results (called partition nodes). The nodes are the points connecting road segments. Then, at least one road segment group is used to update the initial partitioning result, and the updated initial partitioning result is used as the road network partitioning result.

[0146] Specifically, for each road segment group, it is necessary to first query the historical partition results corresponding to multiple boundary nodes on the road segment group boundary in the historical preprocessed road network data with segmentation results. Based on the different historical partition results, different strategies are adopted to update the initial partition results:

[0147] When the historical partitioning results indicate that multiple boundary nodes are divided into the same first specified partition, that is, the partition node of this group of boundary points has only one partitioning result, then in the initial partitioning results, all road sections included in the road section group are divided into the first specified partition, that is, the new data of the current group is also assigned to this partition.

[0148] When the historical partitioning results indicate that multiple boundary nodes are divided into at least two second designated partitions, it means that there is more than one partitioning result for the partition nodes of the group of boundary points. It is necessary to count the number of nodes of the boundary nodes divided into each second designated partition separately. When the number of nodes of at least two second designated partitions is different, in the initial partitioning result, all the road sections included in the road section group are divided into the second designated partition with the highest number of nodes in at least two second designated partitions, that is, the new data of the current group is assigned to the partition where the partition node with the most touches is located; and when the number of nodes of at least two second designated partitions is the same, in the initial partitioning result, all the road sections included in the road section group are randomly divided into any second designated partition of at least two second designated partitions, that is, if the number of nodes is the same, they are randomly assigned to 1 partition.

[0149] When the historical partitioning results indicate that multiple boundary nodes are not divided into any partitions, it means that there is no partition node for the group of boundary points. In this case, a new independent partition is constructed for the road segment group in the initial partitioning results, and the newly added data of the current group becomes a new independent partition.

[0150] In this way, by identifying the diff of road network data and completing incremental segmentation on this basis, we can ensure the accurate identification and efficient processing of road network data changes, as well as the flexible update strategy of the initial partitioning results. This can achieve incremental segmentation of the road network with a small amount of road network updates, and can obtain reasonable road network partitioning results in a short time, and the results are highly matched with the instant delivery scenario.

[0151] In summary, the logical process of the technical solution proposed in this application is summarized as follows: Figure 2D As shown, after the road network partitioning operation begins, the road network data to be partitioned is preprocessed, and idle road network data is eliminated to obtain preprocessed road network data. Then, instant delivery scenario data matching the road network data to be partitioned is obtained. According to the instant delivery scenario data, the road segment attributes are set for each road segment included in the preprocessed road network data, and the road network data to be partitioned and the instant delivery scenario data are merged. Finally, the preprocessed road network data is partitioned. Partitioning can be performed using either full partitioning or incremental partitioning. When partitioning using the full partitioning method, the preprocessed road network data needs to be pre-segmented to obtain pre-segmented road network data. The pre-segmented road network data is then segmented to obtain pre-segmented road network data with segmentation results. The pre-segmented road network data with segmentation results is then restored to preprocessed road network data with segmentation results, and the preprocessed road network data with segmentation results is used as the road network partitioning result. When using the incremental partitioning method for partitioning, it is necessary to identify the differences in the road network, perform incremental segmentation based on the difference marking results, obtain the road network partitioning results, and end the current partitioning operation.

[0152] The method provided in the embodiment of the present application identifies the differences in road network data and completes incremental segmentation on this basis, thereby ensuring accurate identification and efficient processing of road network data changes, as well as a flexible update strategy for the initial partitioning results. It can realize incremental segmentation of the road network with a small amount of road network updates, and can obtain reasonable road network partitioning results in a short time. The results are highly compatible with the instant delivery scenario, and the rationality and flexibility of the road network partitioning are good, thereby ensuring the accuracy and efficiency of subsequent path planning.

[0153] Further, as Figure 1 In the specific implementation of the method, the embodiment of the present application provides a road network partitioning device, such as Figure 3 As shown, the device includes: a pre-processing module 301, a fusion module 302 and a partitioning module 303.

[0154] The pre-processing module 301 is used to determine the road network data to be partitioned, remove the idle road network data from the road network data to obtain pre-processed road network data;

[0155] The fusion module 302 is used to obtain instant delivery scenario data that matches the road network data to be partitioned, and set a road segment attribute for each road segment included in the pre-processed road network data according to the instant delivery scenario data;

[0156] The partitioning module 303 is used to partition the pre-processed road network data according to the road segment attributes corresponding to each road segment included in the pre-processed road network data to obtain a road network partitioning result.

[0157] In a specific application scenario, the preprocessing module 301 is used to match the historical order data of the road network data to be partitioned within a specified historical time period, aggregate the delivery trajectory data in the historical order data, and determine at least one delivery activity area in the road network data to be partitioned; determine a preset expansion distance, expand the range of each delivery activity area according to the expansion distance, and use the at least one expanded delivery activity area as at least one commonly used road network area; use the road network data located between the at least one commonly used road network area in the road network data to be partitioned as the idle road network data, and eliminate the idle road network data, and use the road network data to be partitioned after elimination as the preprocessed road network data.

[0158] In a specific application scenario, the fusion module 302 is used to obtain the instant delivery scene data that matches the road network data to be partitioned, and the instant delivery scene data includes but is not limited to business district range data, delivery range data, and point of interest area AOI boundary data; perform spatial intersection processing on the instant delivery scene data and the pre-processed road network data, determine at least one first road section that intersects with the instant delivery scene data in the road sections included in the pre-processed road network data, and use a preset intersection identifier to set the road section attribute for the at least one first road section; determine in the pre-processed road network data Determine at least one second road section, and use a preset non-intersecting identifier to set section attributes for the at least one second road section, wherein the at least one second road section is a section other than the at least one first road section in the preprocessed road network data; based on the instant delivery scenario data, count the section flow data of each section included in the preprocessed road network data, and determine the section type data of each section; based on the section flow data and section type data of each section, adjust the section attributes currently set for the section included in the preprocessed road network data to complete the setting of the section attributes.

[0159] In a specific application scenario, the fusion module 302 is used to determine at least one road section to be adjusted among the road sections included in the pre-processed road network data, and read the road section attributes currently set for each road section to be adjusted, wherein the road section flow data corresponding to the at least one road section to be adjusted is lower than a preset flow threshold and the corresponding road section type data belongs to a preset special road section type; determine the first road section to be adjusted among the at least one road section to be adjusted whose road section attributes are currently set to the intersection mark, and adjust the road section attributes of the first road section to be adjusted to the preset special mark; determine the second road section to be adjusted among the at least one road section to be adjusted whose road section attributes are currently set to the non-intersection mark, and adjust the road section attributes of the second road section to be adjusted to the intersection mark.

[0160] In a specific application scenario, the partitioning module 303 is used to perform local road network merging on the pre-processed road network data to obtain pre-segmented road network data, and obtain a preset weight setting strategy, respectively query the weight value corresponding to the road segment attribute of each road segment in the preset weight setting strategy, set the road segment weight for each road segment according to the queried weight value, adopt the preset road segment segmentation algorithm, segment the pre-segmented road network data according to the road segment weight corresponding to each road segment, obtain the pre-segmented road network data with the segmentation result, and restore the pre-segmented road network data with the segmentation result to the pre-processed road network data with the segmentation result. According to the method, the pre-processed road network data with the segmentation result is used as the road network partitioning result; or, the historical road network partitioning result of the road network data to be partitioned is read, and the historical pre-processed road network data with the segmentation result with the smallest time interval between the segmentation time point and the current time point is extracted from the historical road network partitioning result, and the pre-processed road network data and the historical pre-processed road network data with the segmentation result are differentially marked with reference to the road segment attributes corresponding to each road segment included in the pre-processed road network data to obtain a differential marking result, and the pre-processed road network data is partitioned according to the differential marking result to obtain the road network partitioning result.

[0161] In a specific application scenario, the partitioning module 303 is used to randomly select any road section node from the road section nodes included in the pre-processed road network data as the target road section node, perform breadth-first search with the target road section node as the starting point, and when a preset number of specified road section nodes are searched, use the road network composed of the preset number of specified road section nodes as a sub-road network; sort the preset number of specified road section nodes according to the order in which they are searched, and divide the sorted preset number of specified road section nodes into a source node group and a sink node group according to a preset division ratio; according to the source node group and the sink node group, use the maximum flow minimum cut algorithm to divide the sub-road network into two temporary road networks, and use the temporary road network including the target road section node in the two temporary road networks as a local road network, and use the road section connected to the local road network in the pre-processed road network data as the boundary section of the local road network; when the local road network meets the preset merging conditions, calculate the average value of the node coordinates of all road section nodes included in the local road network, and The calculated average coordinates are used as temporary node coordinates, and a temporary road section node is created on the temporary node coordinates, wherein the preset merging conditions include but are not limited to the density of road network nodes and the parameter threshold of the maximum flow minimum cut algorithm; a temporary road section is established between the temporary road section node and the boundary road section, and all road section nodes and all road sections included in the local road network are deleted from the pre-processed road network data to complete the merging operation of the local road network; a road section node is randomly selected from other road section nodes included in the pre-processed road network data as a new target road section node, a breadth-first search is performed again with the new target road section node as the starting point, and a new local road network is re-determined, and a merging operation is performed on the new local road network until there is no selectable road section node in the pre-processed road network data, completing the local road network merging operation of the pre-processed road network data to obtain the pre-split road network data, wherein the other road section nodes are road section nodes in the pre-processed road network data except the target road section node and the preset number of designated road section nodes.

[0162] In a specific application scenario, the partitioning module 303 is used to determine at least one temporary road section node in the pre-segmented road network data with the segmentation results, and query the local road network corresponding to each temporary road section node in the pre-processed road network data, wherein the node coordinates of the temporary road section node are obtained by calculating the average value of the node coordinates of all road section nodes included in the corresponding local road network; the local road network corresponding to each temporary road section node is re-added to the pre-segmented road network data with the segmentation results, and each temporary road section node and the temporary road section established based on each temporary road section node are deleted, and the pre-segmented road network data with the segmentation results after the addition and deletion processing are used as the pre-processed road network data with the segmentation results.

[0163] In a specific application scenario, the partitioning module 303 is used to refer to the road segment attributes corresponding to each road segment included in the preprocessed road network data, compare the preprocessed road network data with the historical preprocessed road network data with the segmentation results, and determine the first difference road segment, the second difference road segment and the third difference road segment in the historical preprocessed road network data with the segmentation results, wherein the two road segment endpoints of the first difference road segment exist in both the historical preprocessed road network data with the segmentation results and the preprocessed road network data and the road segment attributes have not changed, the two road segment endpoints of the second difference road segment exist in both the historical preprocessed road network data with the segmentation results and the preprocessed road network data and the road segment attributes have changed, and the two road segment endpoints of the third difference road segment exist in the historical preprocessed road network data with the segmentation results but do not exist in the preprocessed road network data; perform basic marking on the first difference road segment in the historical preprocessed road network data with the segmentation results, and mark the first difference road segment in the historical preprocessed road network data with the segmentation results. The method further comprises the following steps: marking the second difference road segment as a change in the historical preprocessed road network data with the segmentation result, and marking the third difference road segment as a deletion in the historical preprocessed road network data with the segmentation result, wherein the basic mark is used to indicate that the road segment has not changed, the change mark is used to indicate that the road segment attribute has changed, and the deletion mark is used to indicate that the road segment does not exist; comparing the preprocessed road network data with the historical preprocessed road network data with the segmentation result, determining at least one fourth difference road segment in the preprocessed road network data, and marking the at least one fourth difference road segment as an addition in the preprocessed road network data, wherein the two endpoints of the at least one fourth difference road segment do not exist in the historical preprocessed road network data with the segmentation result but exist in the preprocessed road network data, and the addition mark is used to indicate that the road segment is a newly added road segment; the preprocessed road network data with the completed marking and the historical preprocessed road network data with the segmentation result with the completed marking are used as the difference marking result.

[0164] In a specific application scenario, the partitioning module 303 is used to keep the partitioning result of the first difference road section unchanged on the historical preprocessed road network data with the segmentation result in accordance with the indication of the difference marking result, update the section attribute of the second difference road section, and delete the third difference road section to obtain the updated historical preprocessed road network data with the segmentation result; partition the preprocessed road network data with reference to the updated historical preprocessed road network data with the segmentation result to obtain an initial partitioning result; determine the at least one fourth difference road section in the difference marking result, and divide the at least one fourth difference road section into at least one section group according to the connectivity relationship between the at least one fourth difference road sections; use the at least one section group to update the initial partitioning result, and use the updated initial partitioning result as the road network partitioning result.

[0165] In a specific application scenario, the partitioning module 303 is used to query the historical partitioning results corresponding to the multiple boundary nodes on the boundary of the road section group in the historical pre-processed road network data with the segmentation results for each of the road section groups; when the historical partitioning results indicate that the multiple boundary nodes are divided into the same first specified partition, in the initial partitioning results, all the road sections included in the road section group are divided into the first specified partition; when the historical partitioning results indicate that the multiple boundary nodes are divided into at least two second specified partitions, the number of nodes of the boundary nodes divided into each of the second specified partitions is counted respectively, and in the When the number of nodes of at least two second designated partitions is different, in the initial partitioning result, all the road sections included in the road section group are divided into the second designated partition with the highest number of nodes among the at least two second designated partitions; and when the number of nodes of the at least two second designated partitions is the same, in the initial partitioning result, all the road sections included in the road section group are randomly divided into any second designated partition among the at least two second designated partitions; when the historical partitioning result indicates that the multiple boundary nodes are not divided into any partition, a new independent partition is constructed for the road section group in the initial partitioning result.

[0166] The device provided in the embodiment of the present application determines the road network data to be partitioned, eliminates the idle road network data in the road network data to be partitioned, obtains pre-processed road network data, obtains instant delivery scenario data matching the road network data to be partitioned, sets section attributes for each road section included in the pre-processed road network data according to the instant delivery scenario data, partitions the pre-processed road network data according to the section attributes corresponding to each road section included in the pre-processed road network data, and obtains a road network partitioning result. By integrating the instant delivery scenario data, both natural cuts and the characteristics of the instant delivery scenario are taken into account in the road network partitioning process. At the same time, by eliminating idle road network data, the scale of the road network to be partitioned is reduced, and the efficiency of road network partitioning is improved. The rationality and flexibility of road network partitioning are good, thereby ensuring the accuracy and efficiency of subsequent path planning.

[0167] It should be noted that for other corresponding descriptions of the functional units involved in the road network partitioning device provided in the embodiment of the present application, please refer to Figure 1 and Figures 2A to 2D The corresponding description in will not be repeated here.

[0168] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties.

[0169] The above embodiments and the technical features in the embodiments can be arbitrarily combined with each other. In order to make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0170] The above-described embodiments merely represent several implementation methods of the present application. While the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the present application. It should be noted that a person of ordinary skill in the art may make various modifications and improvements without departing from the spirit of the present application, and these modifications and improvements fall within the scope of protection of the present application. Therefore, the scope of protection of the present application shall be determined by the appended claims.

[0171] In an exemplary embodiment, see Figure 4 A computer device is also provided, comprising a bus, a processor, a memory, and a communication interface. The computer device may also include an input / output interface and a display device, wherein the various functional units can communicate with each other via the bus. The memory stores a computer program, and the processor is configured to execute the program stored in the memory and perform the road network partitioning method described in the above embodiment.

[0172] A computer-readable storage medium stores a computer program, which implements the steps of the road network partitioning method when executed by a processor.

[0173] Through the description of the above implementation methods, those skilled in the art can clearly understand that the present application can be implemented through hardware or by means of software plus a necessary general hardware platform. Based on this understanding, the technical solution of the present application can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, a USB flash drive, a mobile hard disk, etc.), including a number of instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each implementation scenario of the present application.

[0174] Those skilled in the art will understand that the accompanying drawings are only schematic diagrams of a preferred implementation scenario, and the modules or processes in the accompanying drawings are not necessarily required to implement the present application.

[0175] Those skilled in the art will appreciate that the modules in the devices in the implementation scenario can be distributed in the devices of the implementation scenario according to the implementation scenario description, or can be modified accordingly and located in one or more devices different from the implementation scenario. The modules in the above implementation scenario can be combined into one module or further split into multiple submodules.

[0176] The above application serial numbers are for description only and do not represent the advantages or disadvantages of the implementation scenarios.

[0177] The above disclosure only describes several specific implementation scenarios of the present application. However, the present application is not limited thereto, and any changes that can be conceived by those skilled in the art should fall within the scope of protection of the present application.

Claims

1. A road network partitioning method, characterized in that: include: Determining road network data to be partitioned, removing idle road network data from the road network data to be partitioned, and obtaining preprocessed road network data; Acquire instant delivery scenario data that matches the road network data to be partitioned, and set a road segment attribute for each road segment included in the pre-processed road network data according to the instant delivery scenario data; According to the road segment attributes corresponding to each road segment included in the pre-processed road network data, the pre-processed road network data is partitioned to obtain a road network partition result.

2. The method according to claim 1, characterized in that The step of removing idle road network data from the road network data to be partitioned to obtain pre-processed road network data includes: Matching the road network data to be partitioned with historical order data within a specified historical time period, aggregating delivery trajectory data in the historical order data, and determining at least one delivery activity area in the road network data to be partitioned; Determining a preset expansion distance, expanding the range of each of the delivery activity areas according to the expansion distance, and using at least one expanded delivery activity area as at least one common road network area; The road network data to be partitioned, which are located between the at least one commonly used road network area, are used as the idle road network data, and the idle road network data are eliminated, and the eliminated road network data to be partitioned are used as the preprocessed road network data.

3. The method according to claim 1, characterized in that The acquiring of instant delivery scenario data and setting a road segment attribute for each road segment included in the pre-processed road network data according to the instant delivery scenario data include: Acquire the instant delivery scenario data that matches the road network data to be partitioned, the instant delivery scenario data including but not limited to business district range data, delivery range data, and point of interest area AOI boundary data; Performing spatial intersection processing on the instant delivery scenario data and the pre-processed road network data, determining at least one first road segment intersecting with the instant delivery scenario data from among the road segments included in the pre-processed road network data, and setting a road segment attribute for the at least one first road segment using a preset intersection identifier; Determining at least one second road segment in the pre-processed road network data, and setting a road segment attribute for the at least one second road segment using a preset non-intersection identifier, wherein the at least one second road segment is a road segment other than the at least one first road segment among the road segments included in the pre-processed road network data; According to the instant delivery scenario data, counting the road section flow data of each road section included in the pre-processed road network data, and determining the road section type data of each road section; Based on the road segment flow data and the road segment type data of each road segment, the road segment attributes currently set for the road segments included in the pre-processed road network data are adjusted to complete the setting of the road segment attributes.

4. The method according to claim 3, characterized in that The adjusting of the road segment attributes currently set for the road segments included in the pre-processed road network data based on the road segment flow data and the road segment type data of each road segment comprises: Determining at least one road section to be adjusted from the road sections included in the pre-processed road network data, and reading the road section attributes currently set for each of the road sections to be adjusted, wherein the road section flow data corresponding to the at least one road section to be adjusted is lower than a preset flow threshold and the corresponding road section type data belongs to a preset special road section type; Determining, among the at least one road section to be adjusted, a first road section to be adjusted whose road section attribute is currently set to the intersection identifier, and adjusting the road section attribute of the first road section to be adjusted to a preset special identifier; A second road section to be adjusted whose road section attribute is currently set to the non-intersection mark is determined in the at least one road section to be adjusted, and the road section attribute of the second road section to be adjusted is adjusted to the intersection mark.

5. The method according to claim 1, wherein The performing partitioning processing on the pre-processed road network data according to the road segment attributes corresponding to each road segment included in the pre-processed road network data to obtain a road network partitioning result includes: Performing local road network merging on the pre-processed road network data to obtain pre-segmented road network data, and obtaining a preset weight setting strategy, respectively querying the weight value corresponding to the section attribute of each road section in the preset weight setting strategy, setting a section weight for each road section according to the queried weight value, adopting a preset section segmentation algorithm, segmenting the pre-segmented road network data according to the section weight corresponding to each road section, obtaining the pre-segmented road network data with segmentation results, and restoring the pre-segmented road network data with segmentation results to the pre-processed road network data with segmentation results, and using the pre-processed road network data with segmentation results as the road network partitioning result; or, Read the historical road network partitioning results of the road network data to be partitioned, extract the historical pre-processed road network data with the segmentation results with the smallest time interval between the segmentation time point and the current time point from the historical road network partitioning results, refer to the section attributes corresponding to each road section included in the pre-processed road network data, perform difference marking on the pre-processed road network data and the historical pre-processed road network data with the segmentation results to obtain a difference marking result, and partition the pre-processed road network data according to the difference marking result to obtain the road network partitioning result.

6. The method according to claim 5, characterized in that The performing local road network merging on the pre-processed road network data to obtain pre-segmented road network data includes: Randomly selecting any road section node from the road section nodes included in the pre-processed road network data as a target road section node, performing a breadth-first search with the target road section node as a starting point, and when a preset number of designated road section nodes are found, using the road network formed by the preset number of designated road section nodes as a sub-road network; Sorting the preset number of designated road section nodes according to the order in which they are searched, and dividing the sorted preset number of designated road section nodes into a source node group and a sink node group according to a preset division ratio; Dividing the sub-road network into two temporary road networks using a maximum flow minimum cut algorithm according to the source node group and the sink node group, taking the temporary road network including the target road segment node of the two temporary road networks as a local road network, and taking the road segment connected to the local road network in the pre-processed road network data as a boundary segment of the local road network; When the local road network meets preset merging conditions, average node coordinates of all road segment nodes included in the local road network are calculated, the calculated average coordinates are used as temporary node coordinates, and a temporary road segment node is created at the temporary node coordinates, wherein the preset merging conditions include but are not limited to the density of road network nodes and the parameter threshold of the maximum flow minimum cut algorithm; Establishing a temporary road section between the temporary road section node and the boundary road section, and deleting all road section nodes and all road sections included in the local road network from the pre-processed road network data, thereby completing a merging operation on the local road network; A road segment node is randomly selected from other road segment nodes included in the preprocessed road network data as a new target road segment node, a breadth-first search is performed again with the new target road segment node as the starting point, and a new local road network is re-determined, and a merging operation is performed on the new local road network until there are no selectable road segment nodes in the preprocessed road network data, completing the operation of merging the local road network of the preprocessed road network data to obtain the pre-segmented road network data, wherein the other road segment nodes are the road segment nodes in the preprocessed road network data except the target road segment node and the preset number of designated road segment nodes.

7. The method according to claim 6, characterized in that The step of restoring the pre-segmented road network data with segmentation results to the pre-processed road network data with segmentation results includes: Determining at least one temporary road segment node in the pre-segmented road network data with the segmentation result, and querying the pre-processed road network data for a local road network corresponding to each temporary road segment node, wherein the node coordinates of the temporary road segment node are calculated by averaging the node coordinates of all road segment nodes included in the local road network corresponding to the temporary road segment node; The local road network corresponding to each temporary road section node is re-added to the pre-segmented road network data with the segmentation results, and each temporary road section node and the temporary road section established based on each temporary road section node are deleted, and the pre-segmented road network data with the segmentation results after the addition and deletion processing is used as the preprocessed road network data with the segmentation results.

8. A road network partitioning device, characterized in that: include: A preprocessing module is used to determine the road network data to be partitioned, and remove the idle road network data from the road network data to be partitioned to obtain preprocessed road network data; a fusion module, configured to obtain instant delivery scenario data matching the road network data to be partitioned, and set a road segment attribute for each road segment included in the pre-processed road network data according to the instant delivery scenario data; The partitioning module is used to partition the pre-processed road network data according to the road segment attributes corresponding to each road segment included in the pre-processed road network data to obtain a road network partitioning result.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images