An optimal path analysis method based on dynamic road network information
Through segmentation of parallel computing and centralized caching mechanisms, combined with static and dynamic information, the optimal path analysis problem in the existing technology that is difficult to respond quickly is solved, and the method of quickly finding the optimal path is realized, and the response speed is improved.
Patent Information
- Application Number
- CN202111456613.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2021-12-02
- Publication Date
- 2024-11-26
- Estimated Expiration
- 2041-12-02
AI Technical Summary
The prior art is difficult to respond quickly in intelligent transportation and public travel services and achieve truly optimal path analysis, especially on the basis of taking into account real-time dynamic traffic flow information and road conditions and facilities information.
Using segmented parallel computing and centralized caching mechanisms, static and dynamic information are collected, and the calculation volume is reduced through cache processing, and the optimal path is quickly found. Specific steps include data collection, cache data, segmented parallel computing and cache failure processing, and using caching mechanisms and parallel computing to improve response speed.
Implement the optimal path of the backbone network between regions when the system is idle or initialized, reduce the scale of a single block diagram, improve the search speed of a single block area, give full play to the advantages of server-side computing resources, and improve the optimal path search Response speed.
Smart Images

Figure CN114418165B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of dynamic road network analysis, and more specifically, to an optimal path analysis method based on dynamic road network information. Background Art
[0002] Optimal path analysis is one of the most commonly used and useful spatial analysis functions in GIS. To make the optimal path analysis function really play a role in practice and serve smart transportation and public travel services, the following two things must be done well:
[0003] First, the real-time dynamic traffic flow information and road condition and facility information are considered, and the static optimal path analysis is expanded to a dynamic optimal path analysis with the support of temporal domain.
[0004] The second is to package the optimal path analysis function into an optimal path analysis service, and provide path analysis services to various terminals by making full use of the server's computing resources and the Internet.
[0005] At present, all companies in this industry need to solve the problem of how to improve the speed of response and achieve the optimal path. Summary of the invention
[0006] The present invention overcomes the shortcomings of the prior art, provides an optimal path analysis method based on dynamic road network information, which combines static information and dynamic information data and reduces the amount of calculation and quickly finds the optimal path through corresponding cache processing.
[0007] The technical solution of the present invention is as follows:
[0008] An optimal path analysis method based on dynamic road network information, the specific steps are as follows:
[0009] 101) Data collection step: Collect static and dynamic information of each section on the route. Static information includes the length, road conditions, and geographical location of the section. Dynamic information includes real-time information on road conditions such as accidents, traffic volume, indicator light conditions, and foggy weather.
[0010] 102) Data access step: access the starting address and the end address to be analyzed, and obtain the static information and dynamic information of all sections between the two addresses;
[0011] 103) Data caching step: A centralized caching mechanism is used to cache the data collected in step 101), and each server shares a cache server. When the cache is insufficient, a new cache server is added to solve the problem; that is, the new cache server and the previous cache server form a new large-space cache server;
[0012] 104) Data analysis step: The cached data is processed by split parallel computing. Split parallel computing is to divide the directed graph formed by the entire data into several areas, and then calculate the optimal path pairs between the areas to form an upper-level backbone network; based on the backbone network, when analyzing the optimal path between two points, it is only necessary to first analyze the optimal path from the source point and the end point to the boundary point of the area where they are located, and then combine the optimal path of the backbone between the source area and the end point area to quickly obtain the final optimal path.
[0013] Furthermore, the static information includes weight attribute parameters and extreme value attribute parameters. The weight attribute parameters of the road section are accumulated and superimposed to obtain the weight attribute parameters of the integrated section. The weight attribute parameters include the length of the section, the toll amount of the section, etc.; the extreme value attribute parameters of the section only obtain the maximum or minimum value of one of the sections in the integrated section.
[0014] Furthermore, static information of each section on the path is collected manually on site at regular intervals, and dynamic information is collected manually, by GPS location, or by video surveillance recognition.
[0015] Furthermore, the centralized cache mechanism includes a data construction method;
[0016] The data construction method defines the cache content of the optimal path from one node to the other nodes; if the optimal path to a point has not been found, its point set flag is false; it is implemented by the DijCacheItem class, which has the DijData method, addOneStation method, addToRedSet method, and addToBlueSet method;
[0017] DijData method is used to store the shortest path weight of each vertex;
[0018] The addOneStation method dynamically adds a station to the cache item of a source point and initializes the cache value of the station;
[0019] The addToRedSet method moves a point in the blue point set to the red point set. The points in the red point set are the points for which the optimal path has been found, and the points in the blue point set are the points for which the optimal path has not been found.
[0020] The addToBlueSet method removes points from the red point set and returns them to the blue point set. It is often used in situations where dynamic sites are removed.
[0021] Furthermore, the centralized cache mechanism includes a cache hit method;
[0022] Before performing path data analysis, the cache hit method first searches the cache pool for corresponding cache items based on the index number and weight type parameters of the start address and the end address; if not found, a new cache item is created and added to the cache pool; if found, the optimal path analysis is continued based on this cache item;
[0023] If the optimal path to the target point has been analyzed in the same way in the above cache items, it can be directly hit and returned; otherwise, it is necessary to continue the path analysis in the conventional way based on the existing path analysis until the target point is found; at the same time, the result is added to the original cache as the new cache content;
[0024] If analysis conditions are included when performing path data analysis, the cache mechanism will fail and the analysis can only be performed using conventional path analysis methods.
[0025] Furthermore, the centralized caching mechanism includes a cache invalidation method;
[0026] Cache invalidation method: When the static and dynamic information of the road segments on the path changes, the originally calculated optimal path in the cache will be partially invalidated. In order to reduce the consumption of cache reconstruction, efforts should be made to ensure that only the invalid analysis results are cleared, rather than all the results. The specific processing strategies include increasing edge weights, disconnecting edges, reducing edge weights, adding an edge, and reducing an edge, as follows:
[0027] When the edge weight increases, if the optimal path to the end point of the edge has not been found, there is no need to invalidate the cache; this is because at this time the optimal path of all the red points in the red point set will not pass through this edge, and the change in the weight of this edge will not affect them, so the cache values of these red points are still valid;
[0028] If the end point of the edge has already obtained the optimal path, that is, according to the length increasing rule of the optimal path analysis, the starting point must have also obtained the optimal path value, then the following cache invalidation process needs to be performed:
[0029] For all red points whose weights are less than the "optimal path value of the end point", there is no need to perform cache invalidation processing; this is because the optimal path of these red points does not pass through this edge and is not affected by the change of the edge weight;
[0030] For all red points whose weight is greater than or equal to the "optimal path value of the end point", cache invalidation processing needs to be performed; for these red points, it is actually only necessary to invalidate all red points with the end point of the edge as a direct or indirect predecessor vertex;
[0031] Cutting an edge is equivalent to resetting the weight of the edge to infinity, which can be effectively processed by the optimal path according to the above-mentioned "increasing edge weight" method.
[0032] The edge weight decreases. If the optimal path has not yet been found for the end point of the edge, there is no need to invalidate the cache. If the optimal path has already been found for the end point of the edge, the following cache invalidation process is required:
[0033] Add the optimal path value of the starting point to the reduced weight of the edge to get a "new value" of the path value of the end point;
[0034] For all red points whose weight is less than or equal to the "new value", there is no need to perform cache invalidation processing;
[0035] For all red points whose weight is greater than the "new value", cache invalidation processing is required, and all these red points need to recalculate the path analysis result cache content;
[0036] Add an edge and obtain the starting and ending nodes of the newly added edge. If the weight of the newly added edge is greater than or equal to the weight of any existing edge between the starting and ending nodes, no cache invalidation processing is required. If the weight of the newly added edge is the smallest between the starting and ending nodes, the optimal path invalidation processing needs to be performed according to the above-mentioned "edge weight reduction" method.
[0037] Reducing an edge is equivalent to setting the weight of the edge to infinity, and performing the optimal path effectiveness processing according to the above-mentioned "edge weight increase" method.
[0038] Furthermore, several regions of the directed graph in step 104) need to use an instance. In the case of a large amount of data, the specific creation call is as follows:
[0039] When the system starts or initializes, a certain number of instances are created in advance. These instance objects are in a ready state and can be used at any time. These instances are called "instance pools". When a path analysis task execution thread starts, it can quickly obtain an idle instance directly from the instance pool and lock it to prevent other analysis threads from using it. It then performs path analysis and ensures that the instance is released back to the instance in time after the analysis is completed for use by other analysis threads.
[0040] Compared with the prior art, the present invention has the following advantages:
[0041] The present invention fully considers various types of data information, and combines split parallel computing with a centralized cache mechanism to pre-calculate and cache the optimal path of the backbone network between regions when the system is idle or initialized; it divides the routes in several traffic-intensive areas and occasions where there are few connections between areas into reasonable blocks, reduces the scale of a single block graph, and thus greatly improves the path search speed of a single block area. Moreover, the path search within each single block area can be processed in parallel, giving full play to the advantages of distributed computing of server-side computing resources, and ultimately improving the response speed of the entire optimal path search. BRIEF DESCRIPTION OF THE DRAWINGS
[0042] Figure 1 This is a diagram showing an example of dynamic information collection of the present invention;
[0043] Figure 2 This is a framework diagram of the centralized cache mechanism of the present invention;
[0044] Figure 3 It is a schematic diagram of the structure of the analysis diagram instance pool of the present invention. DETAILED DESCRIPTION
[0045] The embodiments of the present invention are described in detail below, wherein the same or similar reference numerals throughout represent the same or similar elements or elements with similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention and cannot be used as a limitation of the present invention.
[0046] It will be understood by those skilled in the art that, unless otherwise defined, all terms (including technical terms and scientific terms) used herein have the same meaning as those generally understood by those skilled in the art in the field to which the present invention belongs. It should also be understood that terms such as those defined in general dictionaries should be understood to have meanings consistent with the meanings in the context of the prior art, and will not be interpreted with idealized or overly formal meanings unless defined as herein.
[0047] The numbers of the steps mentioned in each embodiment are only for the convenience of description and have no substantial sequence relationship. Different steps in each specific embodiment can be combined in different sequences to achieve the purpose of the invention.
[0048] The present invention is further described below in conjunction with the accompanying drawings and specific embodiments.
[0049] Example:
[0050] As shown in the figure, an optimal path analysis method based on dynamic road network information has the following specific steps:
[0051] 101) Data collection step: Collect static and dynamic information of each section on the route. Static information includes the length, road conditions, geographic location, etc. of the section, and may also include information about the vehicle itself. Dynamic information includes real-time information on road conditions such as accidents, traffic volume, indicator light status, foggy weather, etc.
[0052] Static information includes weight attribute parameters and extreme value attribute parameters. The weight attribute parameters of the road section are accumulated and superimposed to obtain the weight attribute parameters of the integrated section. The weight attribute parameters include the length of the section, the toll amount of the section, etc. The extreme value attribute parameters of the section only obtain the maximum or minimum value of one of the sections after integration. The static information of each section on the path is collected manually on site at regular intervals, and the dynamic information is collected manually, by GPS location, and by video surveillance recognition. That is, the corresponding data is accessed by authorized persons to ensure the accuracy of the data.
[0053] The reason why static and dynamic information are taken into consideration is that the current path analysis function usually does not consider various attribute information closely related to network analysis, but simply takes the shortest path as the optimal path. This is not advisable in practice. For example, when congestion occurs on a road section, the shortest path may not be the optimal path for travelers.
[0054] Even if some path analysis functions consider attribute information, they do not distinguish between the road segment attributes and the segmented event characteristics. In actual situations, the distribution of an event characteristic (such as road surface flatness) is often not necessarily distributed by road segment, that is, the event segmentation and the physical road segment are not necessarily the same. In a practical optimal path analysis system, the inherent differences and connections between them must be carefully considered. For example, some road segments are in disrepair and full of potholes, so that vehicles cannot travel at their original speed on this section, which will reduce the attributes of this road segment. However, this kind of real-time information is currently ignored.
[0055] In addition, the current path analysis function does not take into account the characteristic information of the main vehicle. In practice, different vehicles have different traffic requirements for roads. For example, overloaded vehicles and dangerous goods transport vehicles have special requirements for their routes. Therefore, practical optimal path analysis must take into account the vehicle's own traffic restriction information and other specified path search filter conditions.
[0056] 102) Data access step: access the starting address and the end address to be analyzed, and obtain the static and dynamic information of all sections between the two addresses.
[0057] 103) Data caching step: A centralized caching mechanism is used to cache the data collected in step 101), and each server shares a cache server. When the cache is insufficient, a new cache server is added to solve the problem; that is, the new cache server and the previous cache server form a new large-space cache server.
[0058] The centralized cache mechanism includes data construction method, cache hit method and cache invalidation method.
[0059] The data construction method defines the cache content of the optimal path from one node to the other nodes; if the optimal path to a point has not been found, its point set flag is false; it is implemented by the DijCacheItem class, which has the DijData method, addOneStation method, addToRedSet method, and addToBlueSet method;
[0060] DijData method is used to store the shortest path weight of each vertex;
[0061]
[0062] The addOneStation method dynamically adds a station to the cache item of a source point and initializes the cache value of the station. Specifically, it is Public void addOneStation(short iStateIndex, short[]vexIndexs, RoadArcInfo[]segInfo, double dPercent); iStateIndex: the index of the station in this cache item. If the index does not exist, create it; vexIndexs: the start and end node number index of the edge where the station is located, with an array size of 2; segInfo: the bidirectional arc information of the edge where the station is located, with an array size of 2; dPercent: the percentage of the station located at the edge position.
[0063] The addToRedSet method moves a point in the blue point set to the red point set. The points in the red point set are the points for which the optimal path has been found, and the points in the blue point set are the points for which the optimal path has not been found.
[0064] The addToBlueSet method removes the points in the red point set and returns them to the blue point set. It is often used in the case of dynamic site removal. The specific implementation code can be obtained by those skilled in the art through multiple debugging based on experience, so the above classes will not be described in detail.
[0065] The cache hit method, before performing path data analysis, first search the cache pool for corresponding cache items based on the index number and weight type parameters of the start address and the end address; if not found, create a new cache item and add it to the cache pool; if found, continue to perform optimal path analysis based on this cache item; since each cache item caches the optimal path from a certain source point to all other nodes, the cache item itself is also a large one-dimensional array.
[0066] If the optimal path to the target point has been analyzed in the same way in the above cache items, it can be directly hit and returned; otherwise, it is necessary to continue the path analysis in the conventional way based on the existing path analysis until the target point is found; at the same time, the result is added to the original cache as the new cache content;
[0067] If analysis conditions are attached when performing path data analysis, the cache mechanism will fail and the conventional path analysis method can only be used. Among them, the attached analysis conditions, such as requiring the road section weight limit to be greater than 10 tons, are special requirements and non-conventional data analysis, so they all need to be recalculated and analyzed.
[0068] Cache invalidation method: When the static and dynamic information of the road section on the path changes, the originally calculated optimal path in the cache will be partially invalidated. In order to reduce the consumption of cache reconstruction, efforts should be made to ensure that only the invalid analysis results are cleared, rather than all the results. Specific processing strategies include increasing edge weights, breaking edges, reducing edge weights, adding an edge, and reducing an edge (it should be noted that the edge mentioned below refers to a single directed edge. For an actual road section, it should be decomposed into two directed edges, positive and negative, for two analyses), as follows:
[0069] When the edge weight increases, if the optimal path to the end point of the edge has not been found, there is no need to invalidate the cache; this is because at this time the optimal path of all the red points in the red point set will not pass through this edge, and the weight change of this edge will not affect them, so the cache values of these red points are still valid;
[0070] If the end point of the edge has already obtained the optimal path, that is, according to the length increasing rule of the optimal path analysis, the starting point must have also obtained the optimal path value, then the following cache invalidation process needs to be performed:
[0071] There is no need to perform cache invalidation processing for all red points whose weights are less than the "optimal path value of the end point"; this is because the optimal path of these red points does not pass through the edge and is not affected by the change in the weight of the edge; this can be proved by contradiction. Assuming that the optimal path of a red point includes the edge, the end point of the edge must be included in the sequence of vertices in the path in front of it. According to the law of increasing length, the path analysis value of the red point must be greater than the path analysis value of the end point of the edge, which does not meet the premise, so it is proved.
[0072] For all red points whose weight is greater than or equal to the "optimal path value of the end point", cache invalidation processing is required; for this part of the red points, it is actually only necessary to invalidate all the red points with the end point of the edge as a direct or indirect predecessor vertex. The proof is as follows: let this part of the red point subset be A, and the remaining part of the red point subset be B. After the optimal path value of all red points in the red point subset A is invalidated and recalculated, it will increase regardless of whether its predecessor vertex has changed. Therefore, the red points in the B set cannot have the red points in the A set as the predecessor vertex, otherwise it is not optimal. However, considering that finding only the red point set of part A will increase the complexity of the algorithm (a recursive algorithm is required), in practice, it is often sufficient to simply expand the invalidation range to reduce the complexity of the algorithm implementation.
[0073] Cutting an edge is equivalent to resetting the weight of the edge to infinity, which can be effectively processed by the optimal path according to the above-mentioned "increasing edge weight" method.
[0074] When the edge weight decreases, if the end point of the edge has not yet found the optimal path, there is no need to invalidate the cache; if the end point of the edge has already found the optimal path (that is, according to the increasing length rule of the optimal path analysis, the starting point must have also found the optimal path value), the following cache invalidation process is required:
[0075] Add the optimal path value of the starting point to the reduced weight of the edge to get a "new value" of the path value of the end point;
[0076] There is no need to perform cache invalidation processing for all red points whose weights are less than or equal to the "new value". The proof by contradiction is as follows. This is because if the cache values of these red points need to be cleared and the end point of the edge is used as the direct or indirect predecessor vertex, then its weight ratio is still greater than the "new value", which is inconsistent with the description of this proposition.
[0077] For all red points whose weight is greater than the "new value", cache invalidation processing is required, and all these red points need to recalculate the path analysis result cache content;
[0078] Add an edge and obtain the starting and ending nodes of the newly added edge. If the weight of the newly added edge is greater than or equal to the weight of any existing edge between the starting and ending nodes, no cache invalidation processing is required. If the weight of the newly added edge is the smallest between the starting and ending nodes, the optimal path invalidation processing needs to be performed according to the above-mentioned "edge weight reduction" method.
[0079] Reducing an edge is equivalent to setting the weight of the edge to infinity, and performing the optimal path effectiveness processing according to the above-mentioned "edge weight increase" method.
[0080] 104) Data analysis step: The cached data is processed by split parallel computing. Split parallel computing is to divide the directed graph formed by the entire data into several regions, and then calculate the optimal path pairs between each region to form an upper-level backbone network; based on the backbone network, when analyzing the optimal path between two points, it is only necessary to first analyze the optimal path from the source point and the end point to the boundary point of the region where they are located, and then combine the optimal path between the source area and the end point area to quickly obtain the final optimal path. This method is particularly suitable for occasions where there are several traffic-intensive areas and there are few connections between areas. Through reasonable block division, the scale of a single-block graph can be reduced, thereby greatly improving the path search speed of a single-block area, and the path search in each single-block area can be processed in parallel, so that the advantages of distributed computing of server-side computing resources can be fully utilized, and ultimately the response speed of the entire optimal path search can be improved.
[0081] Several areas of the directed graph require an instance. When the amount of data is huge, the specific creation call is as follows:
[0082] When the system is started or initialized, a certain number of instances are created in advance. These instance objects are in a ready state and can be used at any time. These instances are called "instance pools"; when a path analysis task execution thread is started, it can quickly obtain an idle instance directly from the instance pool, lock it to prevent other analysis threads from using it, and then perform path analysis. After the analysis is completed, ensure that the instance is released back to the instance in time for use by other analysis threads. This instance pool or object pool management framework realizes the sharing of limited high-cost resources to the maximum extent, improving the operating efficiency of the entire system. Its specific main designs include the following:
[0083] Data structure design
[0084]
[0085] How to get graph instance
[0086]
[0087]
[0088] In summary, this scheme fully considers various types of data information, and combines split parallel computing with centralized caching mechanism to pre-calculate and cache the optimal path of the backbone network between regions when the system is idle or initialized; there are several traffic-intensive areas and routes with few connections between regions, which are reasonably divided into blocks and reduce the scale of a single block graph, thereby greatly improving the path search speed of a single block area, and the path search within each single block area can be processed in parallel, giving full play to the advantages of distributed computing of server-side computing resources, and ultimately improving the response speed of the entire optimal path search.
[0089] The above is only a preferred embodiment of the present invention. It should be pointed out that ordinary technicians in this technical field can make several improvements and modifications without departing from the concept of the present invention. These improvements and modifications should also be regarded as within the scope of protection of the present invention.
Claims
1. An optimal path analysis method based on dynamic road network information, characterized in that: The specific steps are as follows: 101) Data collection steps: Collect static and dynamic information of each section on the route. Static information includes the length, road conditions, and geographic location of the section. Dynamic information includes accidents, traffic volume, indicator light status, and real-time information on changes in road conditions in foggy weather. 102) Data access step: access the starting address and the end address to be analyzed, and obtain the static and dynamic information of all sections between the two addresses; 103) Data caching step: A centralized caching mechanism is used to cache the data collected in step 101), and each server shares a cache server. When the cache is insufficient, a new cache server is added to solve the problem; that is, the new cache server and the previous cache server form a new large-space cache server; The centralized caching mechanism includes data construction methods; The data construction method defines the cache content of the optimal path from one node to the other nodes; if the optimal path to a point has not been found, its point set flag is false; it is implemented by the DijCacheItem class, which has the Dij method, addOneStation method, addToRedSet method, and addBlueSet method. DijData method is used to store the shortest path weight of each vertex; The addOneStation method dynamically adds a station to the cache item of a source point and initializes the cache value of the station; The addToRedSet method moves a point in the blue point set to the red point set. The points in the red point set are the points for which the optimal path has been found, and the points in the blue point set are the points for which the optimal path has not been found. The addToBlueSet method removes points from the red point set and returns them to the blue point set. It is often used in situations where dynamic sites are removed. The centralized caching mechanism includes a cache hit method; Before performing path data analysis, the cache hit method first searches the cache pool for corresponding cache items based on the index number and weight type parameters of the start address and the end address; if not found, a new cache item is created and added to the cache pool; If found, the optimal path analysis will continue based on this cache item; If the optimal path to the target point has been analyzed in the same way in the above cache items, it will be directly returned; otherwise, the path analysis will continue based on the existing path analysis until the target point is found; at the same time, the result will be added to the original cache as the new cache content; If analysis conditions are included when performing path data analysis, the cache mechanism will fail and path analysis will be performed; Centralized caching mechanisms include cache invalidation methods; Cache invalidation method: When the static and dynamic information of the road section on the path changes, the originally calculated optimal path in the cache will be partially invalidated. In order to reduce the consumption of cache reconstruction, the partially invalidated analysis results are cleared. The specific processing strategies include increasing the edge weight, disconnecting the edge, reducing the edge weight, adding an edge, and reducing an edge. The details are as follows: As the edge weight increases, if the optimal path has not yet been found for the end point of the edge, there is no need to invalidate the cache, and the cache value of the red dot is valid. If the optimal path has been found for the end point of the edge, that is, according to the length increasing rule of the optimal path analysis, the following cache invalidation process is performed: For all red points whose weight is less than the "optimal path value of the end point", no cache invalidation processing is required; For all red points whose weight is greater than or equal to the "optimal path value of the end point", cache invalidation is performed; When an edge is disconnected, it is equivalent to resetting the weight of the edge to infinity, and the optimal path failure is processed according to the above-mentioned "edge weight increase" method; The edge weight decreases. If the optimal path has not been found for the end point of the edge, there is no need to invalidate the cache. If the optimal path has been found for the end point of the edge, the following cache invalidation process is performed: Add the optimal path value of the starting point to the reduced weight of the edge to get a "new value" of the path value of the end point; For all red points whose weight is less than or equal to "new value", there is no need to perform cache invalidation processing; For all red points whose weight is greater than the "new value", perform cache invalidation processing and recalculate the path analysis result cache content; Add an edge and obtain the starting and ending nodes of the newly added edge; if the weight of the newly added edge is greater than or equal to the weight of any existing edge between the starting and ending nodes, no cache invalidation processing is required; if the weight of the newly added edge is the smallest between the starting and ending nodes, the optimal path invalidation processing is performed according to the above "edge weight reduction" method; Reducing an edge is equivalent to setting the weight of the edge to infinity, and the optimal path failure processing is performed according to the above "edge weight increase" method; 104) Data analysis step: The cached data is processed by split parallel computing, wherein the split parallel computing is to first divide the directed graph formed by the entire data into several regions, calculate the optimal path pairs between the regions, and form an upper-level backbone network; based on the backbone network, the optimal paths from the source point and the end point to the boundary points of the region where they are located are analyzed, and the optimal path of the backbone between the source region and the end point region is combined to obtain the final optimal path.
2. The optimal path analysis method based on dynamic road network information according to claim 1 is characterized in that: Static information includes weight attribute parameters and extreme value attribute parameters. The weight attribute parameters of the road section are obtained by cumulative superposition to obtain the weight attribute parameters of the integrated section. The weight attribute parameters include the length of the section and the toll amount of the section. The extreme value attribute parameters of the section only obtain the maximum or minimum value of one of the sections in the integrated section.
3. The optimal path analysis method based on dynamic road network information according to claim 1 is characterized in that: The static information of each section on the path is collected manually on site at regular intervals, while the dynamic information is collected manually, by GPS location, and by video surveillance recognition.
4. The optimal path analysis method based on dynamic road network information according to claim 1, characterized in that: In step 104), several areas of the directed graph need to use an instance. In the case of a large amount of data, the specific creation call is as follows: When the system is started or initialized, a certain number of instances are created in advance. These instance objects are in a ready state and can be used at any time. These instances are called "instance pools". When a path analysis task execution thread is started, an idle instance is quickly obtained directly from the instance pool, locked, and path analysis is performed. After the analysis is completed, the instance is released back to the instance pool for use by other analysis threads.
