A Method for Dynamically Fragmenting and Distributing In-Vehicle Navigation Maps to Reduce Local Memory

Abstract

This application relates to the technical field of navigation map data processing, and specifically relates to a method for dynamically slicing and distributing in-vehicle navigation maps to reduce local memory, including: slicing the map data of the in-vehicle navigation position to obtain each pre-stored map block, and obtaining the usage frequency vector of the historical map blocks; clustering the road information similarity between each pre-stored map block and each historical map block to obtain the road consistency factor, determining the storability of each pre-stored map block, and extracting each effective map block; obtaining the connection tightness of each historical map block through the difference of the usage frequency vector, and combining the similarity between the road information of each historical map block and the road information of each effective map block to determine the cache effectiveness of each historical map block, and performing cache update on the historical map blocks, and then dynamically slicing and distributing the in-vehicle navigation maps. This application can reduce the occupation of the local memory of the client of the dynamic navigation system and relieve the transmission burden of the in-vehicle server.

Description

Technical Field

[0001] This application relates to the technical field of navigation map data processing, and particularly relates to a method for dynamically slicing and distributing in-vehicle navigation maps to reduce local memory. Background Art

[0002] In-vehicle navigation systems are divided into static navigation systems and dynamic navigation systems according to the way of using information. Among them, static navigation systems cannot timely reflect the traffic status of the current road conditions and the topological structure of the road network, making it impossible to accurately plan the correct vehicle driving route; while dynamic navigation systems are based on wireless communication and are equipped with a cache system for caching map data such as vector maps and road network structures. Through the processing methods of storing, updating, and distributing high-precision map data, information sharing between the vehicle and the server is realized to provide more intelligent driving services.

[0003] In the process of storing, updating, and distributing high-precision map data, the prior art uses the method of slicing the high-precision in-vehicle navigation map to form map blocks, and stores the map blocks in the cache system. However, due to the limited storage capacity of the cache system, in order to reduce the memory of the map blocks and ensure that new map blocks can be stored, the original map blocks need to be replaced. The existing cache system uses the FIFO (First In First Out) cache replacement algorithm, and this replacement method ignores the usage frequency and the most recent usage time of the original map blocks, resulting in the high-frequency used map blocks being easily replaced, making the stability of caching the map blocks poor and affecting the caching effect of the in-vehicle cache system. Summary of the Invention

[0004] In order to solve the above technical problems, this application provides a method for dynamically slicing and distributing in-vehicle navigation maps to reduce local memory to solve the existing problems.

[0005] A method for dynamically slicing and distributing in-vehicle navigation maps to reduce local memory in this application adopts the following technical solutions:

[0006] An embodiment of this application provides a method for dynamically slicing and distributing in-vehicle navigation maps to reduce local memory, including the following steps:

[0007] Slice the map data at the current in-vehicle navigation position to obtain each pre-stored map block, and obtain the usage times of all historical map blocks before the current moment, and construct the usage frequency vector of each historical map block;

[0008] Obtain the similarity degree of the road information between each pre-stored map block and each historical map block and perform clustering division. According to the deviation degree between the elements in each cluster and the center point of the cluster, obtain the road consistency factor of each cluster corresponding to each pre-stored map block at the current moment. Combine the average level of the elements in each cluster to determine the storage availability of each pre-stored map block at the current moment. Perform threshold segmentation on the storage availability of all pre-stored map blocks to extract each effective map block;

[0009] Obtain the tightness degree of each historical map block at the current moment through the difference between the usage frequency vectors of each historical map block and the average level of the usage frequency vectors of each historical map block. Combine the similarity degree of the road information between each historical map block and each effective map block to determine the cache effectiveness of each historical map block at the current moment;

[0010] Perform cache update on the historical map blocks through the cache effectiveness of the historical map blocks, and then dynamically distribute the vehicle navigation map in slices.

[0011] Preferably, the construction method of the usage frequency vector is: arrange the daily usage times of each historical map block from the time of storage to the current moment in chronological order to form the usage frequency vector of each historical map block.

[0012] Preferably, the obtaining of the similarity degree of the road information between each pre-stored map block and each historical map block and performing clustering division includes:

[0013] For each pre-stored map block, use the road name, road intersection information, and traffic sign information in the pre-stored map block as the input of the bag-of-words model to extract the road information vector of the pre-stored map block at the current moment. Calculate the similarity degree between the road information vector of the pre-stored map block and the road information vectors of each historical map block, and perform clustering on the similarity degree to obtain each cluster corresponding to the pre-stored map block.

[0014] Preferably, the elements in each cluster corresponding to the pre-stored map block are the similarity degrees between the pre-stored map block and each historical map block regarding the road information vector, and the similarity degree is the Jaccard similarity.

[0015] Preferably, the calculation method of the road consistency factor of each cluster corresponding to each pre-stored map block at the current moment is:

[0016] ; In the formula, is the road consistency factor of the jth cluster corresponding to the u-th pre-stored map block at the current moment, is the exponential function with the natural constant as the base, is the number of elements in the j-th cluster corresponding to the u-th pre-stored map block at the current moment. is the s-th element in the j-th cluster corresponding to the u-th pre-stored map block at the current moment. is the center point of the j-th cluster corresponding to the u-th pre-stored map block at the current moment.

[0017] Preferably, the calculation method of the storage availability of each pre-stored map block at the current moment is as follows:

[0018] ; where is the storage availability of the u-th pre-stored map block at the current moment, is the number of clusters corresponding to the u-th pre-stored map block at the current moment, is the exponential normalization function, is the mean value of elements in the j-th cluster corresponding to the u-th pre-stored map block at the current moment.

[0019] Preferably, the effective map block is a map block with a storage availability higher than the segmentation threshold.

[0020] Preferably, the calculation method of the connection tightness of each historical map block at the current moment is as follows:

[0021] ; where is the connection tightness of the v-th historical map block at the current moment, is the mean value of the usage frequency vectors of the v-th historical map block at the current moment, is the number of historical map blocks in the vehicle-mounted cache system at the current moment, is the distance between the usage frequency vectors of the v-th historical map block and the i-th historical map block at the current moment, and exp() is the exponential function with the natural constant as the base.

[0022] Preferably, the calculation method of the cache effectiveness of each historical map block at the current moment is as follows:

[0023] ; where is the cache effectiveness of the v-th historical map block at the current moment, is the number of effective map blocks at the current moment, is the Jaccard similarity between the road information vector of the v-th historical map block and the road information vector of the h-th effective map block at the current moment.

[0024] Preferably, the cache update for the segmented historical maps includes: at the current moment, arranging all the segmented historical maps in ascending order of cache validity, and replacing and updating the first H segmented historical maps after the arrangement with all the valid map segments.

[0025] The present application has at least the following beneficial effects:

[0026] Through the clustering analysis of the Jaccard similarity of the road information vectors between the pre-stored map segments and the segmented historical maps, the present application measures the road consistency characteristics represented by each clustering cluster, and uses the exponential normalization result of the road consistency factor to perform weighted summation on the element means within each clustering cluster, so as to more accurately extract the storable characteristics of the pre-stored map segments, thereby obtaining the valid map segments that need to be stored, which is used to avoid the problem of cache jitter caused by frequently storing redundant map segments in the in-vehicle cache system in the subsequent process, and improve the stability of caching the map segments;

[0027] Furthermore, by analyzing the closeness of the connections between different segmented historical maps and combining the similarity characteristics between the road information within the segmented historical maps and the valid map segments, the present application measures the cache validity of each segmented historical map, which is used to search for the segmented historical maps with poor cache effectiveness and strong replaceability, thereby reducing the risk that the frequently used map segments are easily replaced, improving the stability of caching the map segments in the in-vehicle cache system, and avoiding the adverse effects on the caching effect in the in-vehicle cache system;

[0028] The present application constructs a replacement update vector for the segmented historical maps through the cache validity of each segmented historical map, and realizes the storage of the valid map segments in the in-vehicle cache system through the replacement update vector, avoiding the frequent storage of redundant map segments, and being able to reduce the occupation of the local memory of the client of the dynamic navigation system and relieve the transmission burden of the in-vehicle server. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present application or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the following drawings are only some embodiments of the present application. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0030] Figure 1 It is a flowchart of the steps of a method for dynamically distributing slices of an in-vehicle navigation map provided by the present application to reduce local memory;

[0031] Figure 2 It is a schematic diagram of a map segment provided by the present application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0032] In order to further elaborate on the technical means and effects adopted by this application to achieve the intended invention purpose, the following, in combination with the accompanying drawings and preferred embodiments, details a method for dynamically slicing and distributing in-vehicle navigation maps to reduce local memory, including its specific implementation manner, structure, features, and effects, as follows. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. In addition, the specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0033] Unless otherwise defined, terms such as "including", "comprising", or any other variant thereof are intended to cover non-exclusive inclusion, such that a circuit structure, article, or device including a series of elements not only includes those elements but also other elements not explicitly listed, or further includes elements inherent to such article or device. Without further limitation, an element defined by the statement "including one..." does not exclude the existence of additional identical elements in the article or device including the said element. Additionally, the term "and / or" used herein includes any and all combinations of one or more of the related listed items. All technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the technical field to which this application belongs.

[0034] The following specifically describes the specific solution of a method for dynamically slicing and distributing in-vehicle navigation maps to reduce local memory provided by this application in combination with the accompanying drawings.

[0035] A method for dynamically slicing and distributing in-vehicle navigation maps to reduce local memory provided by an embodiment of this application. Specifically, please refer to Figure 1 .

[0036] The purpose of this embodiment is to analyze the correlation characteristics between the map blocks during vehicle driving and the historical map blocks, consider the usage frequency characteristics of the historical map blocks, perform substitutability analysis on each historical map block, and improve the cache update method in the in-vehicle cache system based on the substitutability analysis results, reduce the risk that frequently used map blocks are easily replaced, improve the stability of the in-vehicle cache system for caching map blocks, and avoid adverse effects on the cache effect in the in-vehicle cache system. Specifically, it includes the following steps:

[0037] Step 1: Split the map data at the current in-vehicle navigation position to obtain each pre-stored map block, and obtain the usage times of all historical map blocks before the current moment, and construct a usage frequency vector for each historical map block.

[0038] While the vehicle is traveling on a certain urban road, the client of the dynamic navigation system obtains the in-vehicle navigation map within the range around the current position of the in-vehicle navigation through in-vehicle GPS positioning. The surrounding range refers to a geographical area: take the longitude of the current position of the in-vehicle navigation at the current moment and latitude , and the geographical area enclosed by the longitude and latitude , where is the longitude change amount, and is the latitude change amount. In this embodiment, the longitude change amount and the latitude change amount

[0039] are respectively set to 10 and 10. Implementers can adaptively select the longitude change amount and the latitude change amount according to the planned route of the actual in-vehicle navigation.

[0040] Furthermore, perform map segmentation on the in-vehicle navigation map corresponding to the current moment. Uniformly select p segmentation longitude lines and q segmentation latitude lines within the longitude range and latitude range of the in-vehicle navigation map respectively. The in-vehicle navigation map is evenly segmented into each map block by all the segmentation longitude lines and segmentation latitude lines, denoted as each pre-stored map block in the in-vehicle navigation map corresponding to the current moment. Among them, the numbers of the segmentation longitude lines and the segmentation latitude lines are respectively set to 50. Implementers can adaptively select according to the size of the in-vehicle navigation map. Figure 2 Specifically, in this embodiment, the schematic diagram of the map block is as shown in Figure 2 , where respectively represent the 1st segmentation longitude line, the (p - 2)th segmentation longitude line, the (p - 1)th segmentation longitude line, and the pth segmentation longitude line, respectively represent the 1st segmentation latitude line, the (q - 2)th segmentation latitude line, the (q - 1)th segmentation latitude line, and the qth segmentation latitude line. Each small square formed by the segmentation longitude lines and the segmentation latitude lines represents each map block.

[0041] At the same time, obtain all the historical map blocks in the in-vehicle cache system at the current moment through the client of the dynamic navigation system, and obtain the daily usage times of each historical map block from the time it was stored to the current moment. The vector formed by arranging the daily usage times of each historical map block from the time it was stored to the current moment in chronological order is denoted as the usage frequency vector of each historical map block in the in-vehicle cache system at the current moment.

[0042] Step 2: Obtain the similarity degree of the road information between each pre-stored map tile and each historical map tile, and perform clustering division. According to the deviation degree between the elements within each clustering cluster and the center point of the clustering cluster, obtain the road consistency factor of each clustering cluster corresponding to each pre-stored map tile at the current moment. Combine the average level of the elements within each clustering cluster to determine the storability of each pre-stored map tile at the current moment. Perform threshold segmentation on the storability of all pre-stored map tiles, and extract each effective map tile.

[0043] In the process of storing, updating, and distributing high-precision map data, it is necessary to dynamically update the map tiles stored in the storage system. However, the existing storage update method uses the FIFO (First In First Out) cache replacement algorithm, which easily causes frequently used map tiles to be replaced, resulting in frequent replacement of cache entries in the cache system and easily causing cache jitter in the in-vehicle cache system. Therefore, in order to avoid the risk of cache jitter in the in-vehicle cache system, it is necessary to perform replaceability analysis on the historical map tiles in the in-vehicle cache system.

[0044] Through each pre-stored map tile in the in-vehicle navigation map corresponding to the current moment, in this embodiment, collect the road name, road intersection information, and traffic sign information within each pre-stored map tile. It should be noted that in the actual application scenario, the implementer can select the specific road information by himself, and this embodiment does not make special restrictions on this. Take the road name, road intersection information, and traffic sign information within the pre-stored map tile as the input of the bag-of-words model, and record the output of the bag-of-words model as the road information vector of each pre-stored map tile at the current moment. Among them, the bag-of-words model is a well-known technology, and the specific process will not be elaborated. Similarly, collect the road information, road intersections, and traffic sign information within each historical map tile at the current moment, and use the bag-of-words model to obtain the road information vector of each historical map tile at the current moment.

[0045] Generally, not all pre-stored map tiles will be stored in the in-vehicle storage system. Instead, only the map tiles with large changes in road information will be updated and stored in the storage system. Therefore, it is necessary to perform storability analysis on all pre-stored map tiles at the current moment.

[0046] Through the above analysis, calculate the similarity between the road information vector of the u-th pre-stored map block and the road information vectors of all historical map blocks. The measurement method of the similarity is not restricted. The implementer can choose Jaccard similarity, Tanimoto coefficient, or Sørensen-Dice coefficient. In this embodiment, Jaccard similarity is selected to measure the similarity between road information vectors. The data set composed of the Jaccard similarities corresponding to the u-th pre-stored map block is input into a clustering algorithm. The clustering algorithm is not restricted. The implementer can choose CURE hierarchical clustering algorithm (Clustering Using REpresentatives), DPC density peak clustering algorithm (density peak clustering), or DBSCAN clustering algorithm (Density-Based Spatial Clustering of Applications with Noise). In this embodiment, the DPC density peak clustering algorithm is selected to cluster the data set composed of Jaccard similarities. In the algorithm, the cut-off distance is selected such that the number of data within a distance less than the cut-off distance around each data accounts for 2% of all the data in the data set. Denote the output of the DPC density peak clustering algorithm as the respective clustering clusters corresponding to the u-th pre-stored map block. Among them, the DPC density peak clustering algorithm is a well-known technology, and the specific process will not be elaborated.

[0047] By performing clustering analysis on the similarities between the road information vectors of each pre-stored map block and the road information vectors of all historical map blocks, clustering clusters at different similarity levels can be mined. If the consistency of the elements within a certain clustering cluster is higher, it indicates that at the current moment, there are multiple historical map blocks whose similarities with the pre-stored map block in terms of road information approach the same, more accurately reflecting the overall similarity characteristics between the pre-stored map block and all historical map blocks in the cache system at the current moment, thereby more accurately performing the storable analysis on each pre-stored map block.

[0048] Furthermore, according to the clustering result of the Jaccard similarity set, combined with the deviation degree between the elements within each clustering cluster and the center point of the clustering cluster, calculate the storable degree of each pre-stored map block at the current moment:

[0049] ; where is the road consistency factor of the j-th clustering cluster corresponding to the u-th pre-stored map block at the current moment, is the exponential function with the natural constant as the base, is the number of elements within the j-th clustering cluster corresponding to the u-th pre-stored map block at the current moment, is the s-th element in the j-th cluster corresponding to the u-th pre-stored map block at the current moment, is the center point of the j-th cluster corresponding to the u-th pre-stored map block at the current moment.

[0050] If the degree of deviation of all elements in a certain cluster from the cluster center element is smaller, and the elements in this cluster are the similarities between the pre-stored map block and the road information vectors of multiple historical map blocks, it represents a higher degree of consistency of the Jaccard similarity between the road information in this cluster, which can reflect the similarity and consistency characteristics between the pre-stored map block and multiple historical map blocks in the cache system at the current moment, and is used to more accurately analyze the storability of each pre-stored map block in the subsequent process. Therefore, in this embodiment, according to the road consistency factor of each cluster corresponding to each pre-stored map block at the current moment, combined with the average level of elements in each cluster, the storability of each pre-stored map block at the current moment is calculated. In this embodiment, the calculation formula is:

[0051] ; where, is the storability of the u-th pre-stored map block at the current moment, is the number of clusters corresponding to the u-th pre-stored map block at the current moment, is the exponential normalization function, is the mean value of the elements in the j-th cluster corresponding to the u-th pre-stored map block at the current moment.

[0052] Since the road consistency factor can reflect the degree of consistency of the Jaccard similarity between the road information in the cluster, using the exponential normalization result of the road consistency factor as the weight for analyzing the storability of each cluster, and weighted summing the mean values of the elements in each cluster, can more accurately analyze the overall similarity characteristics between the pre-stored map block and all historical map blocks in the cache system at the current moment, so as to more accurately measure the storability of each pre-stored map block, which is used to avoid the problem of cache jitter caused by frequently storing redundant map blocks in the subsequent process, and improve the stability of caching map blocks.

[0053] Further, the storability of all pre-stored map tiles at the current moment is used as the input of the Otsu algorithm. The segmentation threshold is obtained by using the Otsu algorithm. The Otsu algorithm is a well-known technology, and its specific process will not be elaborated here. Furthermore, the map tiles corresponding to the storability higher than the segmentation threshold are used as the valid map tiles at the current moment, and the remaining map tiles are recorded as redundant map tiles. Only the valid map tiles are selected for storage in the vehicle storage system to avoid the phenomenon that the cache entries in the vehicle cache system are frequently replaced by redundant map tiles.

[0054] Step 3: Obtain the tightness of connection of each historical map tile at the current moment through the differences between the usage frequency vectors of each historical map tile and the average level of the usage frequency vectors of each historical map tile. Determine the cache validity of each historical map tile at the current moment according to the number of valid map tiles at the current moment, the tightness of connection of each historical map tile, and the similarity of each historical map tile and each valid map tile in terms of road information.

[0055] Analyze the substitutability of each historical map tile in the vehicle cache system by all the valid map tiles at the current moment. If the similarity between the usage frequency vectors of each historical map tile in the vehicle cache system at the current moment is stronger, and the usage frequencies of each historical map tile are at a relatively high level, it can better illustrate that the connection between the used historical map tiles is closer, and the corresponding map tiles should be retained. For example, in vehicle navigation applications, if multiple historical map tiles that are close to each other are used simultaneously multiple times, it means that these historical map tiles belong to the map tiles on the user's frequently traveled routes. Then, the closer the connection between these map tiles is, the less they should be eliminated.

[0056] Through the above analysis, calculate the tightness of connection of each historical map tile at the current moment:

[0057] ; where is the tightness of connection of the v-th historical map tile at the current moment, is the mean value of the usage frequency vector of the v-th historical map tile at the current moment, is the number of historical map tiles in the vehicle cache system at the current moment, is the distance between the usage frequency vectors of the v-th historical map tile and the i-th historical map tile at the current moment. This distance is used to measure the difference between the usage frequency vectors and can be calculated by using the dtw distance or the Euclidean distance. In this embodiment, the dtw distance is used to calculate the difference between the usage frequency vectors.

[0058] It can be understood that the connection tightness reflects the degree of connection tightness between each historical map block and other historical map blocks at the current moment. In vehicle navigation applications, the closer the connection between historical map blocks, the less these historical map blocks should be eliminated, but rather the corresponding historical map blocks should be retained.

[0059] Meanwhile, when performing cache updates on historical map blocks in the in-vehicle cache system, historical map blocks with relatively low similarity to the road information within the valid map blocks should be updated. Otherwise, it belongs to an invalid update of the historical map blocks, which will increase the risk of cache thrashing in the in-vehicle cache system. At the same time, the higher the connection tightness of the historical map blocks at the current moment, the more it can reflect the importance of the historical map blocks. On the contrary, the lower the connection tightness of the historical map blocks at the current moment, the lower the importance of the historical map blocks, and the more such historical map blocks should be cached and updated.

[0060] Through the above analysis, calculate the cache validity of each historical map block at the current moment:

[0061] ; where is the cache validity of the v-th historical map block at the current moment, is the number of valid map blocks at the current moment, is the Jaccard similarity between the road information vector of the v-th historical map block and the road information vector of the h-th valid map block at the current moment.

[0062] The cache validity reflects the effectiveness characteristics of each historical map block in the in-vehicle cache system at the current moment. If the cache effectiveness characteristic of a certain historical map block at the current moment is smaller, it indicates that the replaceability of the historical map block is stronger. Caching and updating the historical map blocks with poor cache effectiveness in the in-vehicle cache system can reduce the risk that frequently used map blocks are easily replaced, improve the stability of the in-vehicle cache system for caching map blocks, and avoid adverse effects on the caching effect in the in-vehicle cache system.

[0063] Step 4: Perform cache updates on historical map blocks based on the cache validity of historical map blocks, and then dynamically distribute slices of the in-vehicle navigation map.

[0064] In this embodiment, for the convenience of understanding and expression, the vector composed of all historical map blocks at the current moment in ascending order of cache validity is denoted as the replacement and update vector of all historical map blocks at the current moment. Each element in the replacement and update vector represents each historical map block. The closer the historical map block is to the head position of the vector in the replacement and update vector, the stronger the replaceability of the historical map block.

[0065] Furthermore, each valid map tile at the current moment is stored in the vehicle-mounted cache system, the number H of valid map tiles at the current moment is counted, the cache indexes of the first H historical map tiles in the replacement update vector are obtained through the vehicle-mounted cache system, and the first H historical map tiles in the replacement update vector are deleted in the vehicle-mounted cache system through the cache indexes. The H valid map tiles at the current moment are cached on the vehicle-mounted cache system, so as to implement the cache update of the first H historical map tiles in the replacement update vector corresponding to the current moment.

[0066] In vehicle navigation applications, the map search engine retrieves the cached and updated map tiles in the vehicle-mounted cache system according to the route planned by the user driving the vehicle, retrieves the map tiles passed by the route planned by the user driving the vehicle, and the dynamic navigation system splices the map tiles passed by the route planned by the user driving the vehicle, so as to obtain the complete high-precision map required for vehicle driving, and dynamically distributes the complete high-precision map required for vehicle driving to the vehicle navigation application driving the vehicle according to the real-time position of the vehicle, thereby implementing the dynamic sharding and distribution task of the vehicle-mounted navigation map, and being able to reduce the occupation of the local memory of the client and relieve the transmission burden of the vehicle-mounted transmission server.

[0067] It can be understood that the reference to "one embodiment" or "some embodiments" etc. described in the specification of this application means that specific features, structures or characteristics described in connection with this embodiment are included in one or more embodiments of this application. Thus, when "in one embodiment", "in some embodiments", "in other some embodiments", "in still other embodiments" etc. appear in different places in this specification, they do not necessarily all refer to the same embodiment, but mean "one or more but not all embodiments", unless otherwise specifically emphasized in other ways. The terms "comprising", "including", "having" and their variants all mean "including but not limited to", unless otherwise specifically emphasized in other ways.

[0068] It should be noted that the above sequence of the embodiments of this application is only for description and does not represent the superiority or inferiority of the embodiments. And the above specific embodiments of this specification have been described. In addition, the processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous. At the same time, the size of the sequence numbers of the steps in the embodiments does not mean the sequence of execution, and the execution sequence of each process should be determined according to its function and internal logic, and should not constitute any limitation to the implementation process of the embodiments in this specification.

[0069] The above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit it; although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that: they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements on some of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present application, and should all be included within the protection scope of the present application.

Claims

1. A method for dynamically slicing and sending a vehicle navigation map to reduce local memory, characterized in that: The following steps are involved: The map data of the vehicle navigation position at the current moment is divided into pre-stored map blocks, and the usage counts of all historical map blocks before the current moment are obtained. The usage counts of each historical map block from the time of storage to the current moment are arranged in chronological order to form a usage frequency vector of each historical map block; Obtain the similarity between the road information of each pre-stored map block and each historical map block and perform clustering. According to the degree of deviation between the elements in each cluster and the center point of the cluster, obtain the road consistency factor of each cluster corresponding to each pre-stored map block at the current moment. Combined with the average level of the elements in each cluster, determine the storability of each pre-stored map block at the current moment, perform threshold segmentation on the storability of all pre-stored map blocks, and extract each valid map block. The connection density of each historical map block at the current moment is obtained through the difference between the usage frequency vectors of each historical map block and the average level of the usage frequency vector of each historical map block. The cache validity of each historical map block at the current moment is determined according to the number of valid map blocks at the current moment, the connection density of each historical map block, and the similarity of road information between each historical map block and each valid map block; The cache of the historical map blocks is updated through the cache validity of the historical map blocks, and then the dynamic fragments of the vehicle navigation map are issued.

2. A method for dynamically slicing and sending a vehicle navigation map to reduce local memory as claimed in claim 1, characterized in that: The obtaining of the similarity between the road information of each pre-stored map block and each historical map block and performing clustering division includes: For each pre-stored map block, the road name, road intersection information and traffic sign information in the pre-stored map block are used as the input of the bag-of-words model to extract the road information vector of the pre-stored map block at the current moment, calculate the similarity between the road information vector of the pre-stored map block and the road information vector of each historical map block, and cluster the similarities to obtain the cluster clusters corresponding to the pre-stored map block.

3. A method for dynamically slicing and sending a vehicle navigation map to reduce local memory as claimed in claim 2, characterized in that: The elements in each cluster corresponding to the pre-stored map block are the similarities between the pre-stored map block and each historical map block regarding the road information vector, and the similarity is the Jaccard similarity.

4. The method for dynamically slicing and sending a vehicle navigation map to reduce local memory as claimed in claim 1, characterized in that: The calculation method of the road consistency factor of each cluster corresponding to each pre-stored map block at the current moment is: ; In the formula, is the road consistency factor of the jth cluster corresponding to the uth pre-stored map block at the current moment, is an exponential function with a natural constant as base, is the number of elements in the jth cluster corresponding to the uth pre-stored map block at the current moment, is the sth element in the jth cluster corresponding to the uth pre-stored map block at the current moment, It is the center point of the jth cluster corresponding to the uth pre-stored map block at the current moment.

5. The method for dynamically slicing and sending a vehicle navigation map to reduce local memory as claimed in claim 4, characterized in that: The calculation method of the storability of each pre-stored map block at the current moment is: ; In the formula, is the storability of the u-th pre-stored map block at the current moment, is the number of clusters corresponding to the u-th pre-stored map block at the current moment, is the exponential normalization function, It is the mean value of the elements in the jth cluster corresponding to the uth pre-stored map block at the current moment.

6. The method for dynamically slicing and sending a vehicle navigation map to reduce local memory as claimed in claim 1, characterized in that: The valid map blocks are map blocks whose storability is higher than the segmentation threshold.

7. The method for dynamically slicing and sending a vehicle navigation map to reduce local memory as claimed in claim 1, characterized in that: The calculation method of the connection density of each historical map block at the current moment is: ; In the formula, is the connection density of the vth historical map block at the current moment, is the mean of the usage frequency vector of the vth historical map block at the current moment, is the number of historical map blocks in the vehicle cache system at the current moment, is the distance between the usage frequency vectors of the vth historical map block and the ith historical map block at the current moment, and exp() is an exponential function with a natural constant as the base.

8. The method for dynamically slicing and sending a vehicle navigation map to reduce local memory as claimed in claim 2, characterized in that: The calculation method of the cache validity of each historical map block at the current moment is: ; In the formula, is the cache validity of the vth historical map block at the current moment, is the number of valid map blocks at the current moment, is the Jaccard similarity between the road information vector of the vth historical map block at the current moment and the road information vector of the hth valid map block, It is the connection density of the vth historical map block at the current moment.

9. A method for dynamically slicing and sending a vehicle navigation map to reduce local memory as claimed in claim 8, characterized in that: The cache updating of the historical map blocks includes: for the current moment, arranging all the historical map blocks from small to large according to the cache validity, and replacing and updating the first H historical map blocks after the arrangement by all the valid map blocks.

Images