A method for identifying composite interchanges and dividing subinterchange nodes in road networks

Abstract

The present invention proposes a method for identifying composite interchanges in road networks and dividing them into sub-interchange nodes, which includes five steps: data preparation, composite interchange boundary point search and grouping, composite interchange internal node extraction, composite interchange identification and division of different sub-interchanges within the composite interchange, remaining node processing, and generation of sub-interchange node sets. In view of the diverse structural forms and complex node connection relationships of composite interchanges in road networks, the present invention proposes a composite interchange node identification method based on the shortest path search between boundary points of groupings in different directions. Furthermore, different sub-interchange nodes are divided according to the composite interchange node search and identification results of each group. This method can automatically and quickly identify different composite interchanges in a road network and accurately and effectively divide the scope of each sub-interchange. It innovates the interchange identification method system and lays the foundation for the subsequent development of interchange mergers in road traffic networks.

Description

Technical Field

[0001] The present invention belongs to the field of virtual integrated transportation system network basic database optimization, and specifically is a method for identifying composite interchanges in a road network and dividing sub-interchange nodes. Background Art

[0002] With the rapid development of transportation infrastructure construction technology in my country, road infrastructure systems are constantly improving and gradually shifting from the transportation planning and construction phase to the transportation demand management phase. This has placed higher demands on the organization and optimization of road traffic operations. Leveraging big data technology to build virtual road traffic systems and simulate road traffic can provide scientific decision-making advice for traffic operation organization, traffic resource allocation, and traffic policy implementation. Therefore, building a comprehensive virtual road traffic system plays a vital role in improving the efficiency of transportation systems and the traffic environment.

[0003] The core and primary component of a virtual road traffic system is the construction of a road network. Nodes and road segments within a road network are abstract topological representations of real-world traffic entities, encompassing a rich array of basic traffic data. Interchanges, as common transportation infrastructure within the road network, possess complex internal topological connections and a diverse array of forms and types. However, when conducting traffic demand and operational analysis using a virtual road traffic system, the connectivity within the interchange is not considered, focusing solely on the direction transitions between different road segments outside the interchange. Furthermore, the large number of nodes and road segments stored within the interchange increases the computational and storage burden of the virtual road traffic system. Compared to conventional interchanges, composite interchanges within interchanges have more complex structures and forms, typically consisting of two or more conventional interchanges. For these composite interchanges, in addition to focusing on the direction transitions between road segments outside the interchange, it is also necessary to clarify the connectivity between the different sub-interchanges within the composite interchange. Therefore, to further optimize the road network infrastructure and alleviate operational pressures on the virtual road traffic system, rapidly identifying composite interchanges within the road network and accurately classifying their sub-interchanges is crucial.

[0004] Current methods for identifying interchanges in road networks focus on identifying typical interchanges found within the network, primarily through manual processing and interchange feature matching. Manual processing involves manual effort to identify various interchange types within a road network. While this method is suitable for identifying and classifying the various sub-interchange components within a complex interchange and offers high accuracy, it is time-consuming and inefficient. Interchange feature matching establishes a library of typical interchange patterns and provides relevant metrics and parameters for each, such as curvature, parallelism, and section density. Interchanges within the network are then identified by calculating metrics for each component within the network and matching these metrics against the library. This method is limited by the types of typical interchanges in the library and is therefore unsuitable for complex interchanges with diverse structures and complex connections. This patent proposes a method for identifying and classifying sub-interchange nodes within complex interchanges within a road network based on the shortest path search between boundary points grouped in different directions. This provides a practical new solution for rapidly identifying complex interchanges within a road network and accurately classifying their sub-interchanges. This method is used to identify and divide the nodes of different sub-interchanges in each composite interchange in the road network, laying the foundation for the subsequent merging of road network interchange nodes.

[0005] The main differences between this application and the prior art are as follows:

[0006] The main comparative differences: The path planning method and system mainly study the improvement of the fast expansion random tree algorithm to extract the driving path of the mobile device, and search for the path by continuously executing the loop to determine whether the distance between the expansion node and the target point is less than the set threshold. In practical applications, the number of loop iterations and the computational efficiency depend on the size of the map, and it is impossible to quickly implement path planning in a large-scale map network. This patent mainly studies the rapid positioning of compound interchanges under existing road networks, the identification of nodes within compound interchanges, and the precise division of different sub-interchanges within compound interchanges. The proposed shortest path search method between boundary points of different direction groups can determine whether there is an intersection of nodes on each path according to the search path of each group, thereby dividing different sub-interchanges, and taking into account the remaining nodes that have not been identified within the range of the compound interchange, and designing a method for processing the remaining nodes based on the divided sub-interchanges, ultimately realizing the identification of all nodes in each compound interchange of the road network and the division of their sub-interchanges.

[0007] 1. Technical comparison with patent CN116069013A "A path planning method and system"

[0008] 1. The path planning method proposed in patent CN116069013A aims to address the low efficiency of existing path planning and the difficulty of routes meeting the travel needs of mobile devices. The main research objectives are to improve the target search efficiency of the rapidly expanding random tree algorithm and reduce redundant nodes in the path search. In contrast, the present invention aims to solve the problem of identifying nodes in complex interchanges and dividing them into different sub-interchanges in road networks. The main research objectives are to quickly identify nodes in each complex interchange in the road network and accurately divide the complex interchanges into different sub-interchanges.

[0009] 2. Patent CN116069013A proposes a path planning method and system that enables rapid search of planned paths within a preset set of map nodes and path expansion within the node safety range. In contrast, the present invention uses a shortest path algorithm to perform a path search on the boundary points of a composite interchange grouped in different directions to identify nodes within the composite interchange. The node sets searched in each direction are further divided into subinterchanges within the composite interchange, and the remaining unidentified nodes are processed based on the distance calculation between the subinterchange midpoints.

[0010] The present invention is mainly used in the field of road network structure optimization.

[0011] 2. Technical comparison with patent CN113808424A "Method for obtaining K shortest paths in urban road network based on bidirectional Dijkstra"

[0012] 1. Patent CN113808424A proposes a shortest path acquisition method that uses a bidirectional search algorithm to obtain k shortest paths between any starting and ending points in an urban road network. The method dynamically selects a complete path by intermittently calculating the sum of the minimum distances in the two currently covered search directions. In contrast, the present invention uses the Dijkstra shortest path algorithm to search for paths between boundary points, after extracting the boundary points of a composite interchange, thereby identifying nodes within the composite interchange. This method can search for paths between boundary points and extract points on the paths within a finite step size, with high operational efficiency and intuitive output.

[0013] 2. Patent CN113808424A proposes a path acquisition method designed to address the low computational efficiency and high resource consumption of traditional network reconstruction and edge deletion methods in the K-shortest path problem. Its research focuses on improving the Dijkstra algorithm for K-shortest path acquisition. In contrast, the method for identifying composite interchanges and dividing subinterchange nodes in a road network proposed in the present invention aims to rapidly identify different composite interchanges in a road network and accurately divide them into subinterchanges. In this invention, the Dijkstra shortest path algorithm is primarily used to search for paths between boundary points to identify nodes within a composite interchange. Summary of the Invention

[0014] To address the aforementioned technical issues, this paper proposes a method for identifying composite interchanges and dividing them into subinterchange nodes within a road network. The method comprises the following steps: data preparation, composite interchange boundary point search and grouping, composite interchange internal node extraction, composite interchange internal node identification and division, residual node processing, and the generation of subinterchange node sets. This method can specifically identify composite interchanges composed of multiple closely spaced interchanges within a road network and effectively divide them into their subinterchanges, laying the foundation for road network optimization and the merging of redundant interchange nodes.

[0015] To achieve the above object, the technical solution adopted by the present invention is:

[0016] A method for identifying composite interchanges and dividing sub-interchange nodes in a road network, comprising the following specific steps:

[0017] (1) Data preparation;

[0018] This step includes two steps: basic network generation and potential interchange set generation. The potential interchange set obtained in step (1) is the basis for composite interchange identification and subinterchange division.

[0019] The basic network information is generated by reading the number of network nodes and road sections, the node types in the network, the node interchange function attributes in the network, the node numbers and coordinates in the network, the connection relationship between each node and the road section attribute information in the network to generate the basic network;

[0020] The generation of potential interchange sets is to divide the mainline bifurcation points in the network into multiple clusters based on the interchange range threshold. Since the distance between nodes of different interchanges in the road network is large, while the distance between nodes within the same interchange is small, the distance between each mainline bifurcation point is calculated and judged whether it is less than the interchange range threshold, and then the different mainline bifurcation points are divided into multiple potential interchange sets.

[0021] (2) Searching and grouping boundary points of composite interchanges;

[0022] The alpha shape algorithm is used to search for the intersection boundary point sets of each potential intersection set in the road network. Furthermore, since a compound intersection is composed of two or more conventional intersections, it is necessary to group the compound intersection boundary point sets into pairs according to different directions and determine the starting and ending boundary points within each group. This lays the foundation for subsequent node identification and sub-interchange division within the compound intersection.

[0023] (3) Extraction of internal nodes of composite interchange;

[0024] Since all points on the path from one boundary point of the interchange to another boundary point belong to the nodes within the range of the interchange, the paths between the boundary points of the same group and across groups are searched for in the different groups in the set of boundary points of each composite interchange obtained in step (2), the interchange nodes of each group are determined, and the path search interchange node set of each group is generated. Subsequently, each sub-interchange is divided according to the path search interchange node set of different groups.

[0025] (4) Identification and division of different sub-interchanges within a complex interchange;

[0026] Each composite interchange in the road network is processed sequentially, and sub-interchanges are divided according to the path search interchange node sets grouped by boundary points in different directions of the composite interchange. That is, if the path search interchange node sets of group 1 and group 2 intersect, they are merged. By judging the path search interchange node sets of each group within the composite interchange, several sub-interchange node sets of the composite interchange are finally obtained.

[0027] (5) Processing of remaining nodes and generation of node sets for each sub-interchange;

[0028] Each composite interchange in the road network is processed in turn. The complete set of composite interchange range nodes is determined by searching the paths between the composite interchange boundary points. The difference between the complete set of composite interchange nodes and the total set of composite interchange sub-interchange nodes obtained by searching in step (4) is obtained to obtain the remaining node set that has not been searched. The midpoints of each sub-interchange are calculated based on the composite interchange sub-interchange node sets. The remaining node sets are traversed and the distances from each remaining node to the midpoints of each sub-interchange are determined in turn. Each remaining node is classified into the sub-interchange set closest to its midpoint. Finally, the complete composite interchange sub-interchange node sets are updated and written into the corresponding composite interchange node file.

[0029] As a further improvement of the present invention, the specific process of step (2) is as follows:

[0030] Step 21, traverse the potential interchanges in all potential interchange sets RInter of the road network k , for each rough k The boundary point set edge of each potential intersection k is obtained by the alpha shape algorithm k , the edge k Stored in the set All_edge of all interchange boundary points of the road network;

[0031] Step 22, select the boundary point set edge of the potential intersection k in All_edge in turn k Processing, if edge k If the number of boundary points is greater than 8, then the potential interchange k is a composite interchange. kThe boundary points in it are grouped in pairs, and the boundary points within each group are the nearest neighbor boundary points to each other, that is, vset m =[n i ,n j , m = 1, 2,.., l, and edge k =[vset1, vset2, …, vset l , at this time, each group vset k in edge m represents the combination of boundary points in a certain direction of the compound interchange; if the number of boundary points in edge k is less than or equal to 8, then the current interchange is not a compound interchange and is not processed, and it is removed from All_edge;

[0032] [[ID=1…]]Step 23, traverse each set of potential interchange boundary points edge k , if the starting boundary point and the ending boundary point of edge k have been discriminated, then select edge k+1 for processing, otherwise, go to Step 24;

[0033] Step 24, sequentially discriminate each direction group vset k in edge m . If the number of boundary points within the group is 1, then do not process; if the number of boundary points within the group is 2, then calculate the distances d1 = dis(v m =[n i ,n j ) of the two boundary points n i ,n j in the group to the midpoint n i ,v j of all boundary points in edge k . If d1 < d2, it means that v p is closer to the midpoint than v i . Therefore, n p is the starting boundary point and n j is the ending boundary point. Otherwise, n p is the ending boundary point and n i is the starting boundary point, and convert this group to vset m =[n j ,n i .

[0034] As a further improvement of the present invention, the specific process of step (3) is as follows: ​​​​​​​​​​

[0035] Step 31, select the boundary point set edge of potential interchange k in turn k Each group in vset m To process, first determine the vset currently being processed m The number of inner boundary points, if it is 2, go to step 3.3; if it is 1, go to step 3.2, if edge k All vsets in m All have been processed, then the edge k+1 to process;

[0036] Step 32, for vset m The only boundary point n in i , traverse the boundary point set edge of potential intersection k k Divide by n i The remaining boundary points n j , if you can search from n within k1 steps i to n j The path is stored in the collection of nodes along the way. middle;

[0037] Step 33, if vset m There are two boundary points in the , the first node is the starting boundary point n ori , the second node is the terminal boundary point n des , if we can search from n within k2 steps ori to n des The path is stored in the collection of nodes along the way. If no path is found, go to step 34;

[0038] Step 34, traverse the intersection boundary point set edge k Divide by n ori and n des The remaining boundary points n j , if we can search from n within k3 steps ori to n j The path is stored in the collection of nodes along the way. middle.

[0039] As a further improvement of the present invention, the specific process of step (4) is as follows:

[0040] This step will process the search node set searchNode(k) of the composite interchange k in turn. For searchNode(k), if any two elements If there is an intersection, merge them into a subintersection set sInter a,a=1,2,…,p。 The elements between searchNode(k) are judged in sequence, and finally p sub-interchanges sInter can be formed a , each sub-interchange sInter a It includes the nodes belonging to the same sub-interchange, and finally these sub-interchanges are stored in the set ComInter(k), which represents the total set of sub-interchange nodes of the composite interchange k.

[0041] As a further improvement of the present invention, the specific process of step (5) is as follows:

[0042] Step 51, traverse the boundary point set edge of the composite interchange k k , select two boundary points n from them i and n j , if we can search from n within k4 steps i to n j The path of each point is stored in the set Node(k), and the edge k The path search between any two boundary points in the ,finally obtained set Node(k) is the complete set of nodes within the range of intersection k;

[0043] Step 52, calculating the difference set pendNode(k) between the complete set of nodes Node(k) of the composite interchange k and the complete set of sub-interchange nodes ComInter(k) of the composite interchange k obtained by the path search. This set stores the remaining nodes in the composite interchange k that have not been searched.

[0044] Step 53: Determine the center point of each sub-interchange and select the sub-interchange set sInter in ComInter(k) in turn. a , if sInter a If the center point has been determined, select sInter a+1 Otherwise, go to step 54;

[0045] Step 54, calculate sInter a The average value of the x-coordinate and y-coordinate of each point in mid ,y mid , remember sInter a The coordinate midpoint is and will Store the midpoint set midNode(k) of each sub-interchange of the composite interchange k. For the p sub-interchange sets of the composite interchange k, a total of p sub-interchange midpoints are generated, which are

[0046] Step 55: Determine the subinterchange to which each remaining node belongs, and traverse each remaining node in pendNode(k) in turn to calculate the remaining nodes. To the center point of each sub-interchange in midNode(k) The distance dis a ,but Belong to min(dis a ) and store the point in the sub-interchange set sInter a ;

[0047] Through the above steps (1) to (5), the node set ComInter(k) of each composite interchange k in the road network is finally obtained. ComInter(k) stores the node set of each sub-interchange, ComInter(k) = [sInter1, sInter2, ..., sInter p ],sInter1=[n1,n2,…,n k ].

[0048] As a further improvement of the present invention, the single boundary point path search step length k1, the double boundary point path search step length k2, the cross-boundary point group path search step length k3 in step (3), and the full set of composite interchange node search step length k4 in step (5).

[0049] Compared with the prior art, the present invention provides a method for identifying composite interchanges and dividing sub-interchange nodes in a road network, which has the following advantages:

[0050] This invention innovatively designs a method for searching and grouping boundary points in composite interchanges. All searched boundary points are divided into several groups of two boundary points, based on the distances between them. The starting and ending boundary points of each group are then determined by calculating the distances between the adjacent points of two boundary points within the group and the midpoints of all boundary points. This method, with clear logic and concise steps, enables rapid search and grouping of boundary points in different directions. This method provides data preparation for composite interchange node identification and subinterchange demarcation. Subsequent subinterchange division within the composite interchange will be performed based on the path search results for boundary points in different direction groups.

[0051] Taking into account the varying number of boundary points in different composite interchanges within a road network, the present invention proposes a composite interchange node identification method based on the shortest path search between boundary points grouped in different directions. Three scenarios are designed: single boundary point path search, dual boundary point path search, and cross-boundary point group path search. This method can rapidly identify nodes within composite interchanges in a road network based on the connectivity of nodes within the composite interchange. Furthermore, considering that paths obtained by searching boundary points grouped in different directions belonging to the same sub-interchange inevitably contain identical nodes, a composite interchange sub-interchange division method is proposed. Based on the path search results for boundary points grouped in each direction, it is determined whether the nodes on each path intersect, thereby dividing the sub-interchanges into different sub-interchanges. Compared to interchange feature matching methods, the proposed method is more applicable, independent of the selection and matching of feature indicators, and not limited by the structural form of the composite interchange. It can rapidly identify nodes within composite interchanges in large-scale road networks and accurately divide the composite interchange into its sub-interchanges.

[0052] This paper designs a method for processing remaining nodes within a defined subinterchange. By calculating the distance from unidentified remaining nodes to the midpoint of each subinterchange, the remaining nodes are classified into their respective subinterchanges. This method, with clear logic and efficient operation, can identify all nodes within each complex interchange in a road network and effectively divide them into subinterchanges. This method lays the foundation for subsequent road network structural optimization. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] Figure 1 is a flow chart of the method of the present invention;

[0054] Figure 2 It is a flow chart for generating a set of potential interchanges;

[0055] Figure 3 It is a schematic diagram of the boundary points of the composite interchange;

[0056] Figure 4 It is a schematic diagram for distinguishing the starting boundary point and the ending boundary point within the group;

[0057] Figure 5 This is a schematic diagram of the internal node extraction of the composite interchange;

[0058] Figure 6 This is a schematic diagram of the sub-interchange division of a composite interchange;

[0059] Figure 7 This is a schematic diagram of searching the entire set of composite interchange nodes;

[0060] Figure 8 It is a schematic diagram for determining the sub-interchange to which each remaining node belongs;

[0061] Figure 9 is a road network structure diagram of an embodiment;

[0062] Figure 10 is a grouping diagram of a potential interchange boundary point in an embodiment network;

[0063] Figure 11 This is a diagram showing preliminary identification and division results of sub-interchanges of the network composite interchange 1 of the embodiment;

[0064] Figure 12 is the remaining node graph of the network composite interchange 1 of the embodiment;

[0065] Figure 13 This is a midpoint diagram of each sub-interchange generated by the network composite interchange 1 of the embodiment;

[0066] Figure 14 This is a result diagram of each sub-interchange obtained after processing the remaining nodes of the network composite interchange 1 of the embodiment;

[0067] Figure 15 This is a diagram showing the results of identifying each composite interchange node and dividing it into sub-interchanges in the network of the embodiment. DETAILED DESCRIPTION

[0068] The technical solution of the present invention will be described in detail below with reference to the accompanying drawings.

[0069] The present invention designs a method for identifying composite interchanges in road networks and dividing sub-interchange nodes. This method can specifically identify composite interchanges in road networks consisting of multiple interchanges with similar distances and effectively divide the sub-interchanges therein, laying the foundation for road network optimization and merging redundant interchange nodes. Figure 1 The figure shows the overall flow chart of the method for identifying composite interchanges in a road network and dividing them into sub-interchange nodes. The steps are as follows:

[0070] Step 1: Data Preparation. This step includes two steps: generating a basic network and generating a potential interchange set. The potential interchange set obtained in Step 1 is the basis for identifying composite interchanges and dividing them into subinterchanges.

[0071] Basic network information is generated primarily by reading information such as the number of network nodes and road segments, node types, node interchange functional attributes, node numbers and coordinates, the connectivity between nodes, and road segment attributes. Node interchange functional attributes include four types: mainline bifurcations, ramps, entry / exit points, and invalid points. Mainline bifurcations connect the mainline of a highway to ramps, ramps connect only ramps, entry / exit points connect national and provincial highways to ramps, and invalid points connect other roads.

[0072] The generation of potential interchange sets is to divide the mainline bifurcation points in the network into multiple clusters according to the interchange range threshold. The specific process is as follows: Figure 2As shown. Since the distance between nodes of different interchanges in the road network is large, while the distance between nodes in the same interchange is small. Therefore, the distance between any two main line bifurcations in the network is calculated in sequence. i and n j The distance d ij , and judge whether it is consistent with the interchange range threshold d r If d ij <d r , then these two nodes belong to the same interchange and are put into the same set rough k After processing all the mainline bifurcations, several potential interchange sets can be obtained. k , k=1,2,3,…n, each rough k All the main line bifurcations in the interchange are included. k Stored in the set of all potential interchanges in the road network RInter

[0073] Step 2: Search and group the boundary points of composite interchanges. k After that, it is necessary to search for the boundary points of the composite interchange. At the same time, since the composite interchange is composed of two or more conventional interchanges, it is necessary to group the boundary points of different directions in the composite interchange into two groups and determine the starting and ending boundary points in each group, laying the foundation for the subsequent node identification and sub-interchange division within the composite interchange. The specific steps are as follows:

[0074] Step 21, traverse the rough of all potential interchange sets RInter in the road network k , for each rough k The boundary point set edge of each potential intersection k is obtained by the alpha shape algorithm k . k Stored in the road network all interchange boundary point set All_edge.

[0075] Step 22, select the boundary point set edge of the potential intersection k in All_edge in turn k Processing. k If the number of boundary points is greater than 8, then the potential interchange k is a composite interchange, such as Figure 3 As shown. k The boundary points in the vset are grouped in pairs, and the boundary points in each group are each other's nearest neighbor boundary points, that is, vset m =[n i ,m j ],m=1,2,..,l. And edge k =[vset1,vset2,…,vset l, at this time, edge k each grouping vset in m represents the combination of boundary points in a certain direction of the composite interchange; if edge k the number of boundary points is less than or equal to 8, then the current interchange is not a composite interchange and is not processed, and it is removed from All_edge.

[0076] Step 23, traverse each potential interchange boundary point set edge k . If edge k has been discriminated for the starting boundary point and the ending boundary point, then select edge k+1 for processing. Otherwise, go to Step 24.

[0077] Step 24, sequentially discriminate each direction grouping vset k in edge m . If the number of boundary points in the grouping is 1, it is not processed; if the number of boundary points in the grouping is 2, as Figure 4 shown, calculate the distances d1 = dis(v m =[n i , n j within the two boundary points n i , n j to the midpoint n i , v j to all boundary points in edge k respectively, d1 = dis(v p , n i , n p ), d2 = dis(v j , n p ). If d1 < d2, it means that v i is closer to the midpoint than v j , so n i is the starting boundary point and n j is the ending boundary point. Otherwise, n i is the ending boundary point and n j is the starting boundary point, and convert this grouping to vset m =[n j , n i .

[0078] Perform boundary point grouping and discrimination of the starting boundary point and the ending boundary point for each potential interchange boundary point set edge k in the road network, and then carry out node search and recognition within the composite interchange based on this.

[0079] Step 3: Extract internal nodes of composite interchange. Therefore, in order to effectively distinguish the nodes to which each sub-interchange belongs, the boundary point set edge of each composite interchange k in All_edge is respectively k , group each direction into vset m Perform path search between boundary points of the same group and across groups to obtain the interchange nodes of each group, such as Figure 5 As shown in the figure. The subsequent steps will divide each sub-interchange according to the node search results of different groups. The specific implementation steps are as follows:

[0080] Step 31, select the boundary point set edge of potential interchange k in turn k Group vset in each direction m To process, first determine the vset currently being processed m The number of inner boundary points. If it is 2, go to step 33; if it is 1, go to step 32. k All vsets in m All have been processed, then the edge k+1 to be processed.

[0081] Step 32, for vset m The only boundary point n in i , traverse the boundary point set edge of potential intersection k k Divide by n i The remaining boundary points n j If you can search from n within k1 steps i to n j The path is stored in the collection of nodes along the way. middle.

[0082] Step 33, if vset m There are two boundary points in the , the first node is the starting boundary point n ori , the second node is the terminal boundary point n des If you can search from n within k2 steps ori to n des The path is stored in the collection of nodes along the way. If no path is found, go to step 34.

[0083] Step 34, traverse the intersection boundary point set edge k Divide by n ori and n des The remaining boundary points n j , if we can search from n within k3 steps ori to n j The path is stored in the collection of nodes along the way. middle.

[0084] On the edge k All vsets in m After performing path search to identify nodes within the interchange, the obtained Store them in the search node set searchNode(k) of the composite interchange k in sequence, and store searchNode(k) in the set All_search.

[0085] Step 4: Identify and divide different sub-interchanges within the composite interchange. This step will process the search node set searchNode(k) of the composite interchange k in turn. Figure 6 As shown, for searchNode(k), if any two elements If there is an intersection, merge them into a subintersection set sInter a ,a=1,2,…,p. The elements between searchNode(k) are judged in sequence, and finally p sub-interchanges sInter are formed. a , each sub-interchange sInter a The nodes belonging to the same subinterchange are included. Finally, these subinterchanges are stored in the set ComInter(k), which represents the total set of subinterchange nodes of composite interchange k.

[0086] Step 5: Process the remaining nodes and generate the node sets of each subinterchange. After processing the nodes in the composite interchange that can be found through path search, some nodes are still not found. Therefore, the remaining nodes need to be classified into the corresponding subinterchanges in turn. a Finally, the complete set of sub-interchange nodes of the composite interchange is updated and generated. The composite interchange k in the road network is processed in turn. The specific implementation steps are as follows:

[0087] Step 51, as Figure 7 As shown, traverse the boundary point set edge of the composite interchange k k , select two boundary points n from them i and n j , if we can search from n within k4 steps i to n j The path of each point is stored in the set Node(k). k The path search between any two boundary points in the ,finally obtained set Node(k) is the complete set of nodes within the range of intersection k.

[0088] Step 52: Calculate the difference set pendNode(k) between the complete set of k nodes of the composite interchange Node(k) and the set of k nodes of the composite interchange ComInter(k) obtained by the path search. This set stores the remaining nodes in the composite interchange k that have not been searched.

[0089] Step 53, determine the center point of each sub-interchange. Select the sub-interchange set sInter in ComInter(k) in turn a , if sInter a If the center point has been determined, select sInter a+1 Otherwise, go to step 54. If the center points of all sub-interchange sets have been determined, go to step 55.

[0090] Step 54, calculate sInter a The average value of the x-coordinate and y-coordinate of each point in mid ,y mid , remember sInter a The coordinate midpoint is and will Store the midpoint set midNode(k) of each subinterchange of compound interchange k. For the p subinterchange sets of compound interchange k, a total of p subinterchange midpoints are generated, which are

[0091] Step 55: Determine the subinterchange to which each remaining node belongs. Traverse each remaining node in pendNode(k) in turn, such as Figure 8 As shown, calculate the remaining nodes To the center point of each sub-interchange in midNode(k) The distance dis a ,but Belong to min(dis a ) and store the point in the sub-interchange set sInter a .

[0092] Through the above steps 1 to 5, we can finally get the total set of sub-interchange nodes ComInter(k) of each composite interchange k in the road network. ComInter(k) stores the node set of each sub-interchange. p ],sInter1=[n1,n2,…,n k ].

[0093] Preferably, in the above step 1, the basic network information that needs to be clarified includes the following 6 categories: Category 1, the total number of nodes and the total number of sections in the current network; Category 2, the node types in the network; Category 3, the functional attributes of node interchanges in the network, including 4 categories: main line bifurcation points, ramp points, entry and exit points, and invalid nodes, represented by numbers 1-4; Category 4, the node numbers and relative coordinates in the network. The network is in a relative two-dimensional coordinate system. The positions of all nodes in the network are represented in the form of (x, y), where x represents the horizontal coordinate of the node and y represents the vertical coordinate of the node; Category 5, the connection relationship between each node, including the node number, the number of other nodes connected to the current node, and the number of other connected nodes; Category 6, the attribute information of sections in the network, including the node numbers of the starting and ending points of the section, the section type, the section grade, the section length, the section speed limit, the number of lanes, etc. There are 6 types of road sections, represented by numbers 21-26, namely expressways (represented by number 21), national roads (represented by number 22), provincial roads (represented by number 23), county roads (represented by number 24), rural roads (represented by number 25), and other roads (represented by number 26).

[0094] This method uses multiple parameters, including the interchange range threshold d in step 1. r In step 3, the single boundary point path search step length k1, the double boundary point path search step length k2, the cross-boundary point group path search step length k3, and the full set of composite interchange node search step length k4 in step 5, the above parameters need to be set according to the form and size of the composite interchange.

[0095] In order to verify the effectiveness of the method of the present invention, a specific example of the present invention is provided below.

[0096] The data preparation in step 1 generates a basic road network and determines the potential interchange set; the basic road network contains 1921 nodes and 4267 road sections. The basic road network geometry is as follows: Figure 9 Some nodes are shown in Table 1 below, and some road sections are shown in Table 2 below.

[0097] Table 1 Partial nodes

[0098] Node number Node Type Interchange functional attributes Node x coordinate Node y coordinate 1849 Highway Node Main line bifurcation point 749223.7952 3565054.724 1850 Highway Node Ramp point 739506.446 3581199.771 1851 Highway Node Invalid node 740355.2891 3580627.725 1852 Highway Node Main line bifurcation point 739612.2595 3580561.69 1853 Highway Node Main line bifurcation point 740624.6618 3587390.698 1854 Highway Node Entry and exit points 741239.5705 3586895.539 1855 Highway Node Entry and exit points 740567.7076 3586967.664 1856 Highway Node Entry and exit points 739492.4464 3595970.622 1857 Highway Node Ramp point 739993.7877 3597021.373 1858 Highway Node Invalid node 740135.6456 3596186.213 1859 Highway Node Ramp point 794387.7397 3574851.886

[0099] Table 2 Part of the road sections

[0100]

[0101]

[0102] After generating the basic road network, set the interchange range threshold d rThe threshold is 1000. Based on this threshold, the mainline bifurcation points in the road network are clustered to generate multiple potential interchange sets. The potential interchange sets are shown in the following table.

[0103] Table 3 Potential interchange set

[0104]

[0105] Through step 2, search for the boundary points of each potential composite interchange, group the boundary points, and determine the starting and ending boundary points. Figure 10 Taking the potential interchange in Figure 1 as an example, the boundary points are grouped as [2654, 2650], [702, 2644], [1017, 570], [2680, 2175], [1788, 2642], and [146, 578]. The first element in each group is the starting boundary point, and the second element is the ending boundary point. Therefore, the boundary point set edge1 of Interchange 1 = [[2654, 2650], [702, 2644], [1017, 570], [2680, 2175], [1788, 2642], [146, 578]]. The black dashed lines in the figure are the boundaries of the potential interchange, and the black solid lines are the actual roads in the road network.

[0106] Set parameters k1=5, k2=8, k3=4 respectively, search for the paths between the boundary points of each composite interchange k in All_edge through step 3, and extract the internal node set All_search of all composite interchanges in the road network. Figure 10 The internal node set obtained by searching for the composite interchange 1 in is shown in the following table.

[0107] Table 4 Search node set

[0108] Boundary point grouping Searching for a node collection [2654,2650] [2654,2360,2495,2650] [702,2644] [702,2644,2496,2495,2650,2654] [146,578] [146,578,655,650,619,1733,2642,1788] [1788,2642] [1788,1869,638,641,650,619,1733,2642] [2680,2175] [2680,2175,2246,2300,2322,570,1017] [1017,570] [1017,570,2323,2155,2175,2680]

[0109] Through step 4, each composite interchange is identified and divided into sub-interchanges, and the node set ComInter(k) of each composite interchange is preliminarily obtained.

[0110] in Figure 11 This is the preliminary identification and division of subinterchanges for Composite Interchange 1. Nodes with white centers in black circles are nodes in Subinterchange 1, intersection nodes in black circles are nodes in Subinterchange 2, nodes with black centers in black circles are nodes in Subinterchange 3, and pure black nodes are other nodes. Table 5 shows the preliminary subinterchange node set for Composite Interchange 1.

[0111] Table 5. Composite Interchange 1 Sub-Interchange Node Set comInter(1)

[0112] Sub-interchange number Searching for a node collection 1 [2654,2360,2495,2650,702,2644,2496,2495] 2 [146,578,655,650,619,1733,2642,1788,1869,638,641] 3 [2680,2175,2246,2300,2322,570,1017,2323,2155]

[0113] Set parameter k4=8, process the remaining nodes and generate each sub-interchange node set through step 5, update the node set ComInter(k) of each composite interchange k in the road network, and write it into the file corresponding to the composite interchange node.

[0114] The whole process of step 5 is demonstrated by taking compound interchange 1 as an example. Figure 12 As shown, the complete set of nodes Node(k) of the composite interchange 1 is obtained by searching the paths between any boundary points in step 51. The remaining set of nodes pendNode(k) not found in the above step 4 is determined in step 52. Figure 12 The black cross node in the middle, the node that has been searched is a pure black node. Figure 13 As shown, through step 53 and step 54, the midpoint of each sub-interchange is obtained (black triangle). Through step 55, the distance from each remaining node to the midpoint of each sub-interchange is determined in turn, and it is included in the corresponding sub-interchange. Finally, Figure 14 Shown are the identification results of composite interchange 1 and its subinterchanges.

[0115] The node and sub-interchange division results of each composite interchange in the road network obtained by this method are shown in the following table and Figure 15 shown.

[0116] Table 6 Node identification and sub-interchange division results within each composite interchange of the case road network

[0117]

[0118] The above description is merely a preferred embodiment of the present invention and does not constitute any other form of limitation to the present invention. Any modification or equivalent variation based on the technical essence of the present invention shall still fall within the scope of protection claimed by the present invention.

Claims

1. A method for identifying composite interchanges and dividing sub-interchange nodes in a road network, comprising the following steps, characterized in that: (1) Data preparation; This step includes two steps: basic network generation and potential interchange set generation. The potential interchange set obtained in step (1) is the basis for composite interchange identification and subinterchange division. The basic network information is generated by reading the number of network nodes and road sections, the node types in the network, the node interchange function attributes in the network, the node numbers and coordinates in the network, the connection relationship between each node and the road section attribute information in the network to generate the basic network; The generation of potential interchange sets is to divide the mainline bifurcation points in the network into multiple clusters based on the interchange range threshold. Since the distance between nodes of different interchanges in the road network is large, while the distance between nodes within the same interchange is small, the distance between each mainline bifurcation point is calculated and judged whether it is less than the interchange range threshold, and then the different mainline bifurcation points are divided into multiple potential interchange sets. (2) Searching and grouping boundary points of composite interchanges; The alpha shape algorithm is used to search for the intersection boundary point sets of each potential intersection set in the road network. Furthermore, since a compound intersection is composed of two or more conventional intersections, it is necessary to group the compound intersection boundary point sets into pairs according to different directions and determine the starting and ending boundary points within each group. This lays the foundation for subsequent node identification and sub-interchange division within the compound intersection. (3) Extraction of internal nodes of composite interchange; Since all points on the path from one boundary point of the interchange to another boundary point belong to the nodes within the range of the interchange, the path search between the boundary points of the same group and across groups is performed on the different groups in the composite interchange boundary point set obtained in step 2 to determine the interchange nodes of each group and generate the path search interchange node set of each group. Subsequently, each sub-interchange will be divided according to the path search interchange node set of different groups. (4) Identification and division of different sub-interchanges within a complex interchange; Each composite interchange in the road network is processed sequentially, and sub-interchanges are divided according to the path search interchange node sets grouped by boundary points in different directions of the composite interchange. That is, if the path search interchange node sets of group 1 and group 2 intersect, they are merged. By judging the path search interchange node sets of each group within the composite interchange, several sub-interchange node sets of the composite interchange are finally obtained. (5) Processing of remaining nodes and generation of node sets for each sub-interchange; Each composite interchange in the road network is processed in turn. The complete set of composite interchange range nodes is determined by searching the paths between the composite interchange boundary points. The difference between the complete set of composite interchange nodes and the total set of composite interchange sub-interchange nodes obtained by searching in step (4) is obtained to obtain the remaining node set that has not been searched. The midpoints of each sub-interchange are calculated based on the composite interchange sub-interchange node sets. The remaining node sets are traversed and the distances from each remaining node to the midpoints of each sub-interchange are determined in turn. Each remaining node is classified into the sub-interchange set closest to its midpoint. Finally, the complete composite interchange sub-interchange node sets are updated and written into the corresponding composite interchange node file.

2. The method for identifying composite interchanges and dividing sub-interchange nodes in a road network according to claim 1, characterized in that: The specific process of step (2) is as follows: Step 21, traverse the potential interchanges in all potential interchange sets RInter of the road network k , for each rough k The boundary point set edge of each potential intersection k is obtained by the alpha shape algorithm k , the edge k Stored in the set All_edge of all interchange boundary points of the road network; Step 22, select the boundary point set edge of the potential intersection k in All_edge in turn k Processing, if edge k If the number of boundary points is greater than 8, then the potential interchange k is a composite interchange. k The boundary points in the vset are grouped in pairs, and the boundary points in each group are each other's nearest neighbor boundary points, that is, vset m =[n i ,n j ],m=1,2,..,l, and edge k =[vset1,vset2,…,vset l ], at this time dege k Each group in vset m Represents the boundary point combination in a certain direction of the composite interchange; if edge k If the number of boundary points is less than or equal to 8, the current interchange is not a composite interchange and is not processed and removed from All_edge; Step 23, traverse each potential intersection boundary point set edge k , if edge k If the starting and ending boundary points have been identified, select edge k+1 Process, otherwise, go to step 24; Step 24, successively discriminate each direction group vset k in edge m . If the number of boundary points in the group is 1, no processing is performed; if the number of boundary points in the group is 2, then calculate the distances d1 = dis(v m =[n i ,n j between the two boundary points n i ,n j in the group and the midpoint n i ,v j of all boundary points in edge k . If d1 < d2, it means that v p is closer to the midpoint than v i . Therefore, n p is the starting boundary point and n j is the ending boundary point. Otherwise, n p is the ending boundary point and n i is the starting boundary point. Convert this group to vset j =[n i ,n j .​​​​​​​​​​ 3. The method for identifying composite interchanges and dividing sub-interchange nodes in a road network according to claim 1, characterized in that: The specific process of step (3) is as follows: Step 31, select the boundary point set edge of potential interchange k in turn k Each group in vset m To process, first determine the vset currently being processed m The number of inner boundary points, if it is 2, go to step 3.3; if it is 1, go to step 3.2, if edge k All vsets in m All have been processed, then the edge k+1 to process; Step 32, for vset m The only boundary point n in i , traverse the boundary point set edge of potential intersection k k Divide by n i The remaining boundary points n j , if you can search from n within k1 steps i to n j The path is stored in the collection of nodes along the way. middle; Step 33, if vset m There are two boundary points in the , the first node is the starting boundary point n ori , the second node is the terminal boundary point n des , if we can search from n within k2 steps ori to n des The path is stored in the collection of nodes along the way. If no path is found, go to step 34; Step 34, traverse the intersection boundary point set edge k Divide by n ori and n des The remaining boundary points n j , if we can search from n within k3 steps ori to n j The path is stored in the collection of nodes along the way. middle.

4. The method for identifying composite interchanges and dividing sub-interchange nodes in a road network according to claim 1, characterized in that: The specific process of step (4) is as follows: This step will process the search node set searchNode(k) of the composite interchange k in turn. For searchNode(k), if any two elements If there is an intersection, merge them into a subintersection set sInter a , a=1,2,…,p, judge the elements between searchNode(k) in turn, and finally form p sub-interchanges sInter a , each sub-interchange sInter a It includes the nodes belonging to the same sub-interchange, and finally these sub-interchanges are stored in the set ComInter(k), which represents the total set of sub-interchange nodes of the composite interchange k.

5. The method for identifying composite interchanges and dividing sub-interchange nodes in a road network according to claim 1, characterized in that: The specific process of step (5) is as follows: Step 51, traverse the boundary point set edge of the composite interchange k k , select two boundary points n from them i and n j , if we can search from n within k4 steps i to n j The path of each point is stored in the set Node(k), and the edge k The path search between any two boundary points in the ,finally obtained set Node(k) is the complete set of nodes within the range of intersection k; Step 52, calculating the difference set pendNode(k) between the complete set of nodes Node(k) of the composite interchange k and the complete set of sub-interchange nodes ComInter(k) of the composite interchange k obtained by the path search. This set stores the remaining nodes in the composite interchange k that have not been searched. Step 53: Determine the center point of each sub-interchange and select the sub-interchange set sInter in ComInter(k) in turn. a , if sInter a If the center point has been determined, select sInter a+1 Process, otherwise, go to step 54; Step 54, calculate sInter a The average value of the x-coordinate and y-coordinate of each point in mid ,y mid , remember sInter a The coordinate midpoint is and will Store the midpoint set midNode(k) of each sub-interchange of the composite interchange k. For the p sub-interchange sets of the composite interchange k, a total of p sub-interchange midpoints are generated, which are Step 55: Determine the subinterchange to which each remaining node belongs, and traverse each remaining node in pendNode(k) in turn to calculate the remaining nodes. To the center point of each sub-interchange in midNode(k) The distance dis a ,but Belong to min(dis a ) and store the point in the sub-interchange set sInter a ; Through the above steps 1 to 5, we can finally get the node set ComInter(k) of each composite interchange k in the road network. ComInter(k) stores the node set of each sub-interchange. ComInter(k) = [sInter1, sInter2, ..., sInter p ],sInter1=[n1,n2,…,n k ].

6. The method for identifying composite interchanges and dividing sub-interchange nodes in a road network according to claim 1, characterized in that: The single boundary point path search step length k1, the double boundary point path search step length k2, the cross-boundary point group path search step length k3 in step (3), and the full set of composite interchange node search step length k4 in step (5).

Images