WO2019201135A1 - 基于全局路网特征的数据匹配方法、装置及存储介质 - Google Patents

基于全局路网特征的数据匹配方法、装置及存储介质 Download PDF

Info

Publication number
WO2019201135A1
WO2019201135A1 PCT/CN2019/082132 CN2019082132W WO2019201135A1 WO 2019201135 A1 WO2019201135 A1 WO 2019201135A1 CN 2019082132 W CN2019082132 W CN 2019082132W WO 2019201135 A1 WO2019201135 A1 WO 2019201135A1
Authority
WO
WIPO (PCT)
Prior art keywords
intersection
matching
road network
road
map data
Prior art date
Application number
PCT/CN2019/082132
Other languages
English (en)
French (fr)
Inventor
张程
罗跃军
王璇
Original Assignee
武汉中海庭数据技术有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 武汉中海庭数据技术有限公司 filed Critical 武汉中海庭数据技术有限公司
Publication of WO2019201135A1 publication Critical patent/WO2019201135A1/zh
Priority to US16/702,719 priority Critical patent/US11486714B2/en

Links

Images

Classifications

    • GPHYSICS
    • G01MEASURING; TESTING
    • G01CMEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
    • G01C21/00Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
    • G01C21/26Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 specially adapted for navigation in a road network
    • G01C21/28Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 specially adapted for navigation in a road network with correlation of data from several navigational instruments
    • G01C21/30Map- or contour-matching
    • G01C21/32Structuring or formatting of map data
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06QINFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00Administration; Management
    • G06Q10/04Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
    • G06Q10/047Optimisation of routes or paths, e.g. travelling salesman problem
    • GPHYSICS
    • G01MEASURING; TESTING
    • G01CMEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
    • G01C21/00Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
    • G01C21/26Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 specially adapted for navigation in a road network
    • G01C21/34Route searching; Route guidance
    • G01C21/36Input/output arrangements for on-board computers
    • G01C21/3667Display of a road map
    • G01C21/367Details, e.g. road map scale, orientation, zooming, illumination, level of detail, scrolling of road map or positioning of current position marker
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06NCOMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00Computing arrangements based on biological models
    • G06N3/02Neural networks
    • G06N3/08Learning methods
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06NCOMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00Computing arrangements based on biological models
    • G06N3/02Neural networks

Definitions

  • the present invention relates to, but is not limited to, the field of computer technology and map navigation, and in particular, to a data matching method, device and storage medium based on global road network features.
  • the in-vehicle map must be changed from the traditional use of traditional map data to the combination of traditional map data and high-precision map data.
  • This requires traditional map data to match high-precision map data.
  • traditional maps and high-resolution maps are often different from map providers, this results in different data models and data ranges for traditional maps and high-precision maps, as well as global positioning of the same name for two maps due to data encryption.
  • the coordinates of the system GPS, Global Positioning System
  • time timeliness
  • the matching of the map data in the related art depends on the GPS coordinates, and the GPS coordinates of the same-named map data of different scales are not the same, which cannot match the map data of different scales, and the data model and the data range are different. , can not use a unified method for offset processing.
  • Embodiments of the present invention provide a data matching method, apparatus, and storage medium based on global road network features.
  • intersection groups in map data are used as feature points, and attributes of feature points and feature points are The relative positional relationship is used to match map data of different scales.
  • an embodiment of the present invention provides a data matching method based on a global road network feature, including:
  • the step 1 includes:
  • Step 101 Load traditional map data map1 (ie, first map data), extract intersection points in map1, and obtain adjacent intersection points of each intersection through road connectivity;
  • map1 ie, first map data
  • Step 102 Load high-precision map data map2 (ie, second map data), extract intersection points in map2, and obtain adjacent intersection points of each intersection through road connectivity;
  • map2 ie, second map data
  • Step 103 Construct a intersection group of map1 and map2 through the connectivity relationship of the roads respectively;
  • Step 104 Extracting intersection group attribute information, where the intersection group attribute information includes: connectivity and location relationship between intersection groups, intersection attributes in intersection groups, number of intersections, connectivity and location relationship between intersections within the group, and roads within the group Attribute information and attribute information of the road connecting the intersection group and the intersection group.
  • the attribute information of the road in the group includes: a road name, an administrative division, a road width, a road shape, a road type, and a road type.
  • the step 2 includes:
  • Step 201 assign weights to each attribute of the intersection group according to an experience or a deep learning method
  • Step 202 Taking any intersection group JG i in the traditional map data map1 as an object, searching for the matching intersection group JG′ j in the high-precision map data map2, where i, j are integers;
  • Step 203 If there is a pair of intersection groups in different map data, having the same attribute, the confidence value of the matching of the pair of intersection groups is calculated according to the weight of each attribute; if there is no identical attribute, the matching confidence value is 0;
  • Step 204 Finally, a set of matching intersection groups corresponding to all intersection groups in the traditional map data map1 is obtained.
  • the step 3 includes:
  • Step 301 selecting any intersection group JG i in the traditional map data map1 as an object, and the intersection group matched in the high-precision map is JG' j ;
  • Step 302 if the matched set of adjacent intersection junctions group JG i JG m in the map data with high accuracy in the intersection group map2 JG 'n group is an intersection JG' j and a set of adjacent intersections confidence value is greater than 0, it is considered
  • the road network formed by JG i ⁇ JG m matches the road network formed by JG' j ⁇ JG' n ;
  • Step 303 selecting intersection group JG m for the object, and repeating steps 302, finally obtained matched respectively to the junction group JG i and JG 'j as the center of intersection groups, JG i and JG' all roads communicating relationship j matching configuration Net N i and N' j ;
  • Step 304 For one of the road network N i with the intersection group JG i as the central intersection group, and the plurality of intersection groups in the road network N′ j with the intersection group JG′ j as the central intersection group Confidence values are summed to obtain a confidence value that matches N i and N' j ;
  • Step 305 According to the principle that the intersection and the matching intersection do not cross, all road networks in map1 are obtained, and are recorded as N 0 ... N n (n is an integer) and the corresponding road network combination in map 2 is recorded as N' 0 ... N' n , at this time between N 0 ... N n between two and two and N' 0 ... N' m between the two, there is no matching connected road.
  • Step 306 according to the matching relationship, in map1 and map2, a combination of a plurality of pairs N 0 ... N x (x is an integer, which may be different from n) may be obtained.
  • the road network or intersection group can be repeated, but in the same combination, the intersections included in different road networks or intersection groups and the matching intersections corresponding to the intersections cannot be repeated;
  • Step 307 Calculate the confidence value of the road network combination N 0 ... N n in map1 and the road network combination N' 0 ... N' m in map2 according to the relative positional relationship of each road network in the road network combination , and obtain a confidence.
  • Step 308 taking the road network combination with the largest total value of the correspondence as the matching optimal solution.
  • an embodiment of the present invention further provides a data matching device based on a global road network feature, including:
  • the map loading module is configured to load different map data, respectively construct the intersection group in each map, and extract the attribute information of the intersection group;
  • the confidence value evaluation module is configured to assign weights to each attribute of the intersection group according to the importance degree of the attribute information, and comprehensively evaluate the matching values of the two intersection groups in different maps;
  • the road network matching module is configured to construct a matching road network according to the confidence value, and calculate a total confidence value of all matching road networks in the map, and select a road network combination with the largest total confidence value as a matching optimal solution, that is, a matching result. .
  • the map loading module includes:
  • the first map loading module is configured to load the traditional map data map1, extract the intersection points in the map1, and obtain the adjacent intersection points of each intersection through the road connectivity;
  • the second map loading module is configured to load the high-precision map data map2, extract the intersection points in the map2, and obtain the adjacent intersection points of each intersection through the road connectivity;
  • intersection group building module is configured to respectively construct the intersection groups of map1 and map2 through the connectivity relationship of the roads;
  • the attribute extraction module is configured to extract the attribute information of the intersection group, and the attribute information of the intersection group includes: connectivity and positional relationship between the intersection groups, intersection attributes in the intersection group, number of intersections, connectivity and positional relationship between the intersections within the group, and groups Attribute information of the inner road and attribute information of the road connecting the intersection group and the intersection group.
  • the attribute information of the road in the group includes: a road name, an administrative division, a road width, a road shape, a road type, and a road type.
  • the confidence value evaluation module includes:
  • a weight allocation module configured to assign weights to each attribute of the intersection group according to an experience or a deep learning method
  • the intersection group collection generation module is configured to select any intersection group JG i in the traditional map data map1 as an object, and map2 finds the matching intersection group JG' j in the high-precision map data, wherein i, j are integers;
  • map2 finds the matching intersection group JG' j in the high-precision map data, wherein i, j are integers;
  • a set of matching intersection groups corresponding to all intersection groups in the map data map1.
  • the road network matching module includes:
  • the matching road network generating module is configured to select any intersection group JG i in the traditional map data map1 as an object, and the intersection group matched in the high-precision map is JG' j ; if the intersection group JG i is adjacent to the intersection group JG m In the high-accuracy map data map2, the matching intersection group JG' n is the adjacent intersection group of the intersection group JG' j and the confidence value is greater than 0, then the road network composed of JG i ⁇ JG m is considered to be JG' j ⁇ The road network formed by JG' n matches;
  • the confidence total value calculation module is configured to target one of the road networks N i with the intersection group JG i as the central intersection group, and the plurality of road network groups N′ j with the intersection group JG′ j as the central intersection group
  • the confidence values of the various intersection groups are summed to obtain a confidence value for the matching of N i and N' j .
  • all the road networks in map1 are obtained, which are recorded as N 0 ... N n (n is an integer) and the corresponding road network combination in map2, which is recorded as N' 0 ... N' n , when N 0 ... N n and the communication path between any two N '0 ...
  • N' m between any two did not match up.
  • a combination of a plurality of pairs N 0 ... N x (x is an integer, which can be different from n) can be obtained.
  • the road network or intersection group can be repeated, but in the same combination, the intersections included in different road networks or intersection groups and the matching intersections corresponding to the intersections cannot be repeated.
  • the matching result output module is configured to take the road network combination with the largest total confidence value as the matching optimal solution.
  • the embodiment of the invention further provides a data matching device based on a global road network feature, the device comprising:
  • a memory configured to store data matching programs based on global road network characteristics
  • the processor is configured to run the program, wherein the program is executed to execute the global road network feature-based data matching method provided by the embodiment of the present invention.
  • the embodiment of the present invention further provides a storage medium, where the storage medium includes a stored program, wherein the program is executed to execute the data matching method based on the global road network feature provided by the embodiment of the present invention.
  • High matching rate Due to the feature matching of the overall road network, similar to the data correction in the surveying, there is a sufficient theoretical basis to explain the matching rate problem of the algorithm.
  • the algorithm mainly uses the relative positional relationship, as long as the relative positional relationship of the intersection group in the map is known, the data can be matched, and can be extended to apply to the map data update when positioning between the two maps. The difference.
  • the matching rate can be further improved: the attribute of the intersection group in the algorithm and the confidence value corresponding to the attribute matching can be replaced by the deep learning algorithm, and the matching value and the confidence value corresponding to the attribute matching are continuously optimized, and the matching rate can be matched. Increase to nearly 100%.
  • FIG. 1 is a flowchart of a data matching method based on a global road network feature according to an embodiment of the present invention
  • FIG. 2 is a structural diagram of a data matching apparatus based on a global road network feature according to an embodiment of the present invention.
  • Intersection An area where multiple roads meet, or an area where road rights are selected is called an intersection. The center point of the intersection is recorded as an intersection.
  • Adjacent intersections at intersections refers to the intersections at which the motor vehicle is at the current intersection, through which there is only a passable road (there is no intersection of roads).
  • FIG. 1 is a flowchart of a data matching method based on a global road network feature according to an embodiment of the present invention. As shown in FIG. 1 , the method includes:
  • Step 1 loading different map data, respectively constructing the intersection group in each map, and extracting the attribute information of the intersection group;
  • Step 2 According to the importance degree of the attribute information, assign weights to each attribute of the intersection group, and comprehensively evaluate the matching values of the two intersection groups in different maps;
  • Step 3 Construct a matching road network according to the confidence value, calculate a confidence value of the matched road network, and construct a matching road network combination through the matched road network.
  • the confidence value of the road network combination matching is calculated as the total value of the confidence, and the road network combination with the largest total confidence value is selected as the matching optimal solution, that is, the matching result.
  • the implementation process is as follows:
  • the connectivity relationship refers to the number of roads entering and leaving the intersection (group) and the adjacent intersection information leading to the intersection;
  • the attribute information of the roads in the group includes attributes of roads within the intersection group, including road names, administrative divisions, road widths, road shapes, types of roads, road types, and the like;
  • the point of interest attribute is one or more of the attributes of the intersection group and the attributes of the road between the intersection groups, and the weights are assigned to the points of interest attributes according to experience or deep learning.
  • the traditional map data and the high-precision map data if there is a pair of intersection groups, they have the same focus attribute, and the confidence of the pair of intersection groups can be calculated according to the weight of the focus point. If there is no same focus attribute, the match is 0.
  • the intersection groups JG'1, JG'2, ... JG'n matching JG1 can be found, and are represented by JG'i (i is a value between 1, 2, ... n). If the adjacent intersection group JG11 of JG1 matches the adjacent intersection group JG'ii of JG'i with a confidence value greater than 0, the road network composed of JG1->JG11 and JG'i->JG'ii The road network matches. The adjacent intersection group JG111 of JG11 and the adjacent intersection group JG'iii of JG'ii are found, and the confidence value satisfying the matching of JG111 and JG'iii is greater than zero.
  • the road network composed of JG1->JG11->JG111 matches the road network formed by JG'i->JG'ii->JG'iii.
  • N1 and N'i In order to construct a road network matching JG1 and JG'i, denoted as N1 and N'i;
  • N1 and N'i (the value between i and 1, 2...n) as a new intersection group, and obtain the connection between the road network and the road network in map1 and map2 through the connection relationship of the roads. Relationship and relative location information;
  • the embodiment of the present invention provides a data matching device based on the global road network feature, as shown in FIG. 2, which includes:
  • the map loading module is configured to load different map data, respectively construct the intersection group in each map, and extract the attribute information of the intersection group;
  • the confidence value evaluation module assigns weights to each attribute of the intersection group according to the importance degree of the attribute information, and comprehensively evaluates the matching values of the two intersection groups in different maps;
  • the road network matching module constructs a matching road network according to the confidence value, calculates a confidence value of the matched road network, and constructs a matching road network combination through the matched road network. According to the relative positional relationship of each road network in the matching road network combination, the confidence value of the road network combination matching is calculated as the total value of the confidence, and the road network combination with the largest total confidence value is selected as the matching optimal solution, that is, the matching result.
  • the map loading module includes:
  • the first map loading module is configured to load the traditional map data map1, extract the intersection points in the map1, and obtain the adjacent intersection points of each intersection through the road connectivity;
  • the second map loading module is configured to load the high-precision map data map2, extract the intersection points in the map2, and obtain the adjacent intersection points of each intersection through the road connectivity;
  • intersection group building module is configured to respectively construct the intersection groups of map1 and map2 through the connectivity relationship of the roads;
  • the attribute extraction module is configured to extract the attribute information of the intersection group, and the attribute information of the intersection group includes: connectivity and positional relationship between the intersection groups, intersection attributes in the intersection group, number of intersections, connectivity and positional relationship between the intersections within the group, and groups Attribute information of the inner road and attribute information of the road connecting the intersection group and the intersection group.
  • the attribute information of the road within the group includes: a road name, an administrative division, a road width, a road shape, a type of road, and a road type.
  • the confidence value evaluation module includes:
  • a weight allocation module configured to assign weights to each attribute of the intersection group according to an experience or a deep learning method
  • the intersection group collection generation module is configured to select any intersection group JG i in the traditional map data map1 as an object, and map2 finds the matching intersection group JG' j in the high-precision map data, wherein i, j are integers;
  • map2 finds the matching intersection group JG' j in the high-precision map data, wherein i, j are integers;
  • a set of matching intersection groups corresponding to all intersection groups in the map data map1.
  • the road network matching module includes:
  • the matching road network generating module is configured to select any intersection group JG i in the traditional map data map1 as an object, and the intersection group matched in the high-precision map is JG' j ; if the intersection group JG i is adjacent to the intersection group JG m In the high-accuracy map data map2, the matching intersection group JG' n is the adjacent intersection group of the intersection group JG' j and the confidence value is greater than 0, then the road network composed of JG i ⁇ JG m is considered to be JG' j ⁇ The road network formed by JG' n matches;
  • the confidence total value calculation module is configured to target one of the road networks N i with the intersection group JG i as the central intersection group, and the plurality of road network groups N′ j with the intersection group JG′ j as the central intersection group
  • the confidence values of the various intersection groups are summed to obtain a confidence value for the matching of N i and N' j .
  • all the road networks in map1 are obtained, which are recorded as N 0 ... N n (n is an integer) and the corresponding road network combination in map2, which is recorded as N' 0 ... N' n , when N 0 ... N n and the communication path between any two N '0 ...
  • N' m between any two did not match up.
  • map1 and map2 a combination of a plurality of pairs N 0 ... N x (x is an integer, which can be different from n) can be obtained.
  • the road network or intersection group can be repeated, but in the same combination, the intersections included in different road networks or intersection groups and the matching intersections corresponding to the intersections cannot be repeated.
  • the road network combination N 0 ... N n in map1 is calculated , and the confidence value matching the road network combination N' 0 ... N' m in map2 is obtained, and the total confidence value is obtained. ;
  • High matching rate Due to the feature matching of the overall road network, similar to the data correction in the surveying, there is a sufficient theoretical basis to explain the matching rate problem of the algorithm.
  • the matching rate is improved.
  • the focus attribute and the confidence value corresponding to the attribute matching in the embodiment of the present invention may be replaced by a deep learning algorithm, and the matching point attribute and the confidence value corresponding to the attribute matching are continuously optimized, and the matching may be performed.
  • the rate has increased to nearly 100%.

Abstract

一种基于全局路网特征的数据匹配方法、装置及存储介质,其中方法包括:加载不同地图数据,分别构建各地图中的路口组,并提取路口组属性信息;根据属性信息的重要度,为路口组的各属性分配权重,综合评价不同地图中的两个路口组相匹配的置信值;根据该置信值构建匹配的道路网,并计算匹配的道路网的置信总值,选取置信总值最大的道路网为匹配最优解,即匹配结果。

Description

基于全局路网特征的数据匹配方法、装置及存储介质
相关申请的交叉引用
本申请基于申请号为201810335065.X、申请日为2018年04月15日的中国专利申请提出,并要求该中国专利申请的优先权,该中国专利申请的全部内容在此引入本申请作为参考。
技术领域
本发明涉及但不限于计算机技术及地图导航领域,尤其涉及一种基于全局路网特征的数据匹配方法、装置及存储介质。
背景技术
在自动驾驶新兴的今天,车载地图必然由单纯的使用传统地图数据,变更为传统地图数据与高精度地图数据组合使用的方式。这就需要传统地图数据与高精度地图数据相匹配。但是由于传统地图与高精度地图的地图提供商往往不同,这就导致传统地图与高精度地图的数据模型和数据范围不同,同时也会由于数据加密的原因导致两幅地图的同名点的全球定位系统(GPS,Global Positioning System)坐标不同,甚至偏移很大,而且传统地图与高精度地图之间由于测绘时间不同(时效性),导致路网存在较大的差异。
相关技术中关于地图数据的匹配的都依赖于GPS坐标,而不同尺度的地图数据同名点的GPS坐标并不相同,这就无法对不同尺度的地图数据进行匹配,而且由于数据模型与数据范围不同,无法利用统一的方法进行偏移处理。
发明内容
本发明实施例提供一种基于全局路网特征的数据匹配方法、装置及存储介质,通过提取全局路网特征,将地图数据中的路口组作为特征点,通过特征点的属性以及特征点之间的相对位置关系来进行不同尺度的地图数据的匹配。
一方面,本发明实施例提供一种基于全局路网特征的数据匹配方法,包括:
加载不同地图数据,分别构建各地图中的路口组,并提取路口组属性信息;
根据属性信息的重要度,为路口组的各属性分配权重,综合评价不同地图中的两个路口组相匹配的置信值;
根据所述置信值构建匹配的道路网,并计算匹配的道路网的置信值,再通过匹配的道路网构建匹配道路网组合;根据匹配道路网组合中,各道路网的相对位置关系,计算道路网组合匹配的置信值为置信总值,选取置信总值最大的道路网组合为匹配最优解,即匹配结果。
上述方案中,所述步骤1包括:
步骤101,加载传统地图数据map1(即第一地图数据),提取map1中的路口点,并通过道路连通性获取每个路口的相邻路口点;
步骤102,加载高精度地图数据map2(即第二地图数据),提取map2中的路口点,并通过道路连通性获取每个路口的相邻路口点;
步骤103,通过道路的连通关系分别构建map1和map2的路口组;
步骤104,提取路口组属性信息,所述路口组属性信息包括:路口组之间的连通及位置关系、路口组内路口属性、路口数、组内路口间的连通及位置关系、组内道路的属性信息以及路口组与路口组之间连通道路的属性信息。
上述方案中,所述组内道路的属性信息包括:道路名称、行政区划、道路宽度、道路形状、道路的种别、道路类型。
上述方案中,所述步骤2包括:
步骤201,依据经验或深度学习方法,为路口组的各属性分配权重;
步骤202,取传统地图数据map1中的任一路口组JG i为对象,在高精度地图数据map2中查找相匹配的路口组JG’ j,其中i,j为整数;
步骤203,若在不同的地图数据中,存在一对路口组,具有相同的属性,依据各属性的权重计算这一对路口组匹配的置信值;若没有相同的属性,则匹配的置信值为0;
步骤204,最终得到传统地图数据map1中所有路口组对应的匹配路口组集合。
上述方案中,所述步骤3包括:
步骤301,选取传统地图数据map1中的任一路口组JG i为对象,其在高精度地图中匹配的路口组为JG’ j
步骤302,若路口组JG i的相邻路口组JG m在高精度地图数据map2中的相匹配的路口组JG’ n为路口组JG’ j的相邻路口组且置信值大于0,则认为JG i→JG m构成的道路网与JG’ j→JG’ n构成的道路网相匹配;
步骤303,选取路口组JG m为对象,并重复步骤302,最终得到相匹配的分别以路口组JG i和JG’ j为中心路口组,JG i和JG’ j匹配的所有连通关系构成的道路网N i和N’ j
步骤304,针对以路口组JG i为中心路口组的其中一个道路网N i,对其相匹配的多个以路口组JG’ j为中心路口组的道路网N’ j中的各路口组的置信值求和,得到N i和N’ j相匹配的置信值;
步骤305,根据路口及其匹配路口不交叉的原则,得到map1中所有道路网,记为N 0…N n(n为整数)以及map2中对应的道路网组合,记为N’ 0…N’ n,此时N 0…N n两两之间以及N’ 0…N’ m两两之间,均没有相匹配的连通道路了。
步骤306,根据匹配关系,在map1和map2中,可以得到多对N 0…N x(x为整数,可以与n不同)的组合。在不同的组合中道路网或路口组可以重复,但在同一组合中,不同的道路网或路口组所包含的路口以及路口对应的匹配路口均不能重复;
步骤307,根据道路网组合中各道路网的相对位置关系,计算map1中的道路网组合N 0…N n,与map2中的道路网组合N’ 0…N’ m匹配的置信值,得到置信总值;
步骤308,取置信总值最大的道路网组合为匹配最优解。
另一方面,本发明实施例还提供一种基于全局路网特征的数据的匹配装置,包括:
地图加载模块,配置为加载不同地图数据,分别构建各地图中的路口组,并提取路口组属性信息;
置信值评价模块,配置为根据属性信息的重要度,为路口组的各属性分配权重,综合评价不同地图中的两个路口组相匹配的置信值;
道路网匹配模块,配置为根据所述置信值构建匹配的道路网,并计算地图中所有匹配的道路网的置信总值,选取置信总值最大的道路网组合为匹配最优解,即匹配结果。
上述方案中,所述地图加载模块包括:
第一地图加载模块,配置为加载传统地图数据map1,提取map1中的路口点,并通过道路连通性获取每个路口的相邻路口点;
第二地图加载模块,配置为加载高精度地图数据map2,提取map2中的路口点,并通过道路连通性获取每个路口的相邻路口点;
路口组构建模块,配置为通过道路的连通关系分别构建map1和map2的路口组;
属性提取模块,配置为提取路口组属性信息,所述路口组属性信息包括:路口组之间的连通及位置关系、路口组内路口属性、路口数、组内路 口间的连通及位置关系、组内道路的属性信息以及路口组与路口组之间连通道路的属性信息。
上述方案中,所述组内道路的属性信息包括:道路名称、行政区划、道路宽度、道路形状、道路的种别、道路类型。
上述方案中,所述置信值评价模块包括:
权重分配模块,配置为依据经验或深度学习方法,为路口组的各属性分配权重;
路口组集合生成模块,配置为选取传统地图数据map1中的任一路口组JG i为对象,在高精度地图数据中map2查找相匹配的路口组JG’ j,其中i,j为整数;若在不同的地图数据中,存在一对路口组,具有相同的属性,依据各属性的权重计算这一对路口组匹配的置信值;若没有相同的属性,则匹配的置信值为0;最终得到传统地图数据map1中所有路口组对应的匹配路口组集合。
上述方案中,所述道路网匹配模块包括:
匹配道路网生成模块,配置为选取传统地图数据map1中的任一路口组JG i为对象,其在高精度地图中匹配的路口组为JG’ j;若路口组JG i的相邻路口组JG m在高精度地图数据map2中的相匹配的路口组JG’ n为路口组JG’ j的相邻路口组且置信值大于0,则认为JG i→JG m构成的道路网与JG’ j→JG’ n构成的道路网相匹配;
置信总值计算模块,配置为针对以路口组JG i为中心路口组的其中一个道路网N i,对其相匹配的多个以路口组JG’ j为中心路口组的道路网N’ j中的各路口组的置信值求和,得到N i和N’ j匹配的置信值。根据路口及其匹配路口不交叉的原则,得到map1中所有道路网,记为N 0…N n(n为整数)以及map2中对应的道路网组合,记为N’ 0…N’ n,此时N 0…N n两两之间以及N’ 0…N’ m两两之间,均没有相匹配的连通道路了。根据匹配关系,在map1和map2中,可以得到多对N 0…N x(x为整数,可以与n不同)的组合。在 不同的组合中道路网或路口组可以重复,但在同一组合中,不同的道路网或路口组所包含的路口以及路口对应的匹配路口均不能重复。根据道路网组合中各道路网的相对位置关系,计算map1中的道路网组合N 0…N n,与map2中的道路网组合N’ 0…N’ m匹配的置信值,得到置信总值;
匹配结果输出模块,配置为取置信总值最大的道路网组合为匹配最优解。
本发明实施例还提供了一种基于全局路网特征的数据匹配装置,所述装置包括:
存储器,配置为保存基于全局路网特征的数据匹配的程序;
处理器,配置为运行所述程序,其中,所述程序运行时执行本发明实施例提供的所述基于全局路网特征的数据匹配方法。
本发明实施例还提供了一种存储介质,所述存储介质包括存储的程序,其中,所述程序运行时执行本发明实施例提供的所述基于全局路网特征的数据匹配方法。
应用本发明实施例的有益效果是:
1、匹配率高:由于采用了整体路网的特征匹配,类似于测绘中的数据纠偏,所以有充分的理论基础来说明算法的匹配率问题。
2、可广泛应用:由于算法中主要使用了相对位置关系,那么只要知晓了地图中的路口组的相对位置关系即可进行数据匹配,可延伸应用于地图数据更新时,定位两幅地图之间的差异。
3、可进一步提高匹配率:算法中的路口组属性以及属性匹配所对应的置信值,可以由深度学习算法进行代替,不断优化关注点属性以及属性匹配所对应的置信值,则可以将匹配率提高到接近100%。
附图说明
图1为本发明实施例提供的基于全局路网特征的数据匹配方法流程图;
图2为本发明实施例提供的基于全局路网特征的数据匹配装置结构图。
具体实施方式
以下结合实例对本发明的原理和特征进行描述,所举实例只用于解释本发明,并非用于限定本发明的范围。
首先对本发明实施例中涉及的名词进行解释:
路口:多条道路交汇的区域,或存在路权选择的区域称为路口,路口的中心点记为路口点。
路口的邻接路口:是指机动车在当前路口,通过有且仅有可通行的道路(不存在道路交汇区域),可到达的路口。
接下来以传统地图和高精度地图数据匹配为例对本发明实施例进行说明。
图1为本发明实施例提供的基于全局路网特征的数据匹配方法流程图,如图1所示,包括:
步骤1,加载不同地图数据,分别构建各地图中的路口组,并提取路口组属性信息;
步骤2,根据属性信息的重要度,为路口组的各属性分配权重,综合评价不同地图中的两个路口组相匹配的置信值;
步骤3,根据所述置信值构建匹配的道路网,并计算匹配的道路网的置信值,再通过匹配的道路网构建匹配道路网组合。根据匹配道路网组合中,各道路网的相对位置关系,计算道路网组合匹配的置信值为置信总值,选取置信总值最大的道路网组合为匹配最优解,即匹配结果。
在一些实施例中,实施过程如下:
1、加载传统地图数据map1(即第一地图数据),提取map1中的路口点,并通过道路连通性获取每个路口的相邻路口点;
2、加载高精度地图数据map2(即第二地图数据),提取map2中的路 口,并通过道路连通性获取每个路口的相邻路口;
3、通过道路的连通关系分别构建map1和map2的路口组;
4、提取关注点属性,即路口组之间的连通关系、相对位置关系、路口组内路口属性、路口数、组内路口间的连通关系、相对位置关系,组内道路的属性信息以及路口组与路口组之间连通道路的属性信息等;
所述连通关系是指进入和脱出路口(组)的道路数以及通向的邻接路口信息;
所述组内道路的属性信息包括路口组内的道路的属性,包括道路名称,行政区划,道路宽度,道路形状,道路的种别,道路类型等;
5、以map1中任意路口组JG1在map2中寻找相匹配的路口组JG’并利用关注点属性是否相同,评价其匹配的置信值。此时我们可以找到多个路口组与JG1相匹配,记为JG’1,JG’2……JG’n。
关注点属性为路口组的属性,以及路口组间道路的属性中的一种或多种,并依据经验或深度学习的给这些关注点属性分配权重。传统地图数据和高精度地图数据中,若存在一对路口组,它们有相同的关注点属性时,可依据关注点的权重来计算这一对路口组匹配的置信度。若没有相同的关注点属性,则匹配度为0。
6、根据JG1的匹配结果,可找到与JG1匹配的路口组JG’1,JG’2……JG’n,以JG’i(i为1,2……n之间的数值)表示。若存在JG1的相邻路口组JG11与JG’i的相邻路口组JG’ii相匹配的置信值大于0,则JG1->JG11构成的道路网,与JG’i->JG’ii构成的道路网相匹配。再找到JG11的相邻的路口组JG111,以及JG’ii的相邻路口组JG’iii,满足JG111和JG’iii匹配的置信值大于0。那么JG1->JG11->JG111构成的道路网和JG’i->JG’ii->JG’iii构成的道路网相匹配。以此来构建以JG1与JG’i相匹配的道路网,记为N1和N’i;
7、以N1与N’i中,路口组匹配的置信值求和的方法,分别计算N1 与N’i(i为1,2……n之间的数值)相匹配的置信值;
8、将N1与N’i(i为1,2……n之间的数值)看成新的路口组,通过道路的连通关系,分别获取map1和map2中道路网与道路网之间的连通关系以及相对位置信息;
9、以新的路口组重复第5~8步,直至无法构建新的匹配的道路网为止;
10、获取置信值最大的为匹配最优解。
本发明实施例在上述数据匹配方法的基础上提供一种基于全局路网特征的数据匹配装置,如图2所示,包括:
地图加载模块,配置为加载不同地图数据,分别构建各地图中的路口组,并提取路口组属性信息;
置信值评价模块,根据属性信息的重要度,为路口组的各属性分配权重,综合评价不同地图中的两个路口组相匹配的置信值;
道路网匹配模块,根据所述置信值构建匹配的道路网,并计算匹配的道路网的置信值,再通过匹配的道路网构建匹配道路网组合。根据匹配道路网组合中,各道路网的相对位置关系,计算道路网组合匹配的置信值为置信总值,选取置信总值最大的道路网组合为匹配最优解,即匹配结果。
在一些实施例中,所述地图加载模块包括:
第一地图加载模块,配置为加载传统地图数据map1,提取map1中的路口点,并通过道路连通性获取每个路口的相邻路口点;
第二地图加载模块,配置为加载高精度地图数据map2,提取map2中的路口点,并通过道路连通性获取每个路口的相邻路口点;
路口组构建模块,配置为通过道路的连通关系分别构建map1和map2的路口组;
属性提取模块,配置为提取路口组属性信息,所述路口组属性信息包括:路口组之间的连通及位置关系、路口组内路口属性、路口数、组内路口间的连通及位置关系、组内道路的属性信息以及路口组与路口组之间连 通道路的属性信息。
在一些实施例中,所述组内道路的属性信息包括:道路名称、行政区划、道路宽度、道路形状、道路的种别、道路类型。
在一些实施例中,所述置信值评价模块包括:
权重分配模块,配置为依据经验或深度学习方法,为路口组的各属性分配权重;
路口组集合生成模块,配置为选取传统地图数据map1中的任一路口组JG i为对象,在高精度地图数据中map2查找相匹配的路口组JG’ j,其中i,j为整数;若在不同的地图数据中,存在一对路口组,具有相同的属性,依据各属性的权重计算这一对路口组匹配的置信值;若没有相同的属性,则匹配的置信值为0;最终得到传统地图数据map1中所有路口组对应的匹配路口组集合。
在一些实施例中,所述道路网匹配模块包括:
匹配道路网生成模块,配置为选取传统地图数据map1中的任一路口组JG i为对象,其在高精度地图中匹配的路口组为JG’ j;若路口组JG i的相邻路口组JG m在高精度地图数据map2中的相匹配的路口组JG’ n为路口组JG’ j的相邻路口组且置信值大于0,则认为JG i→JG m构成的道路网与JG’ j→JG’ n构成的道路网相匹配;
置信总值计算模块,配置为针对以路口组JG i为中心路口组的其中一个道路网N i,对其相匹配的多个以路口组JG’ j为中心路口组的道路网N’ j中的各路口组的置信值求和,得到N i和N’ j匹配的置信值。根据路口及其匹配路口不交叉的原则,得到map1中所有道路网,记为N 0…N n(n为整数)以及map2中对应的道路网组合,记为N’ 0…N’ n,此时N 0…N n两两之间以及N’ 0…N’ m两两之间,均没有相匹配的连通道路了。根据匹配关系,在map1和map2中,可以得到多对N 0…N x(x为整数,可以与n不同)的组合。在不同的组合中道路网或路口组可以重复,但在同一组合中,不同的道路网 或路口组所包含的路口以及路口对应的匹配路口均不能重复。根据道路网组合中各道路网的相对位置关系,,计算map1中的道路网组合N 0…N n,与map2中的道路网组合N’ 0…N’ m匹配的置信值,得到置信总值;
匹配结果输出模块,取置信总值最大的道路网为匹配最优解。
应用本发明实施例具备以下有益技术效果:
1、匹配率高:由于采用了整体路网的特征匹配,类似于测绘中的数据纠偏,所以有充分的理论基础来说明算法的匹配率问题。
2、可广泛应用:由于算法中主要使用了相对位置关系,那么只要知晓了地图中的路口组的相对位置关系即可进行数据匹配。还可延伸应用于地图数据更新时,定位两幅地图之间的差异。
3、提高匹配率:本发明实施例中的关注点属性以及属性匹配所对应的置信值,可以由深度学习算法进行代替,不断优化关注点属性以及属性匹配所对应的置信值,则可以将匹配率提高到接近100%。
以上所述仅为本发明的较佳实施例,并不用以限制本发明,凡在本发明的精神和原则之内,所作的任何修改、等同替换、改进等,均应包含在本发明的保护范围之内。

Claims (12)

  1. 一种基于全局路网特征的数据匹配方法,包括:
    加载不同地图数据,分别构建各地图中的路口组,并提取路口组属性信息;
    根据属性信息的重要度,为路口组的各属性分配权重,综合评价不同地图中的两个路口组相匹配的置信值;
    根据所述置信值构建匹配的道路网,并计算匹配的道路网的置信值,再通过匹配的道路网构建匹配道路网组合;根据匹配道路网组合中,各道路网的相对位置关系,计算道路网组合匹配的置信值为置信总值,选取置信总值最大的道路网组合为匹配最优解。
  2. 根据权利要求1所述的方法,其中,所述加载不同地图数据,分别构建各地图中的路口组,并提取路口组属性信息,包括:
    加载第一地图数据,提取所述第一地图数据中的路口点,并通过道路连通性获取每个路口的相邻路口点;
    加载第二地图数据,提取所述第二地图数据中的路口点,并通过道路连通性获取每个路口的相邻路口点;
    通过道路的连通关系分别构建所述第一地图数据和所述第二地图数据的路口组;
    提取路口组属性信息,所述路口组属性信息包括:路口组之间的连通及位置关系、路口组内路口属性、路口数、组内路口间的连通及位置关系、组内道路的属性信息以及路口组与路口组之间连通道路的属性信息。
  3. 根据权利要求2所述的方法,其中,所述组内道路的属性信息包括:道路名称、行政区划、道路宽度、道路形状、道路的种别、道路类型。
  4. 根据权利要求2所述的方法,其中,所述根据属性信息的重要度,为路口组的各属性分配权重,综合评价不同地图中的两个路口组相匹配的 置信值包括:
    为路口组的各属性分配权重;
    取所述第一地图数据中的任一路口组JGi为对象,在所述第二地图数据中查找相匹配的路口组JG’j,其中i,j为整数;
    若在不同的地图数据中,存在一对路口组,具有相同的属性,依据各属性的权重计算这一对路口组匹配的置信值;若没有相同的属性,则匹配的置信值为0;
    得到所述第一地图数据中所有路口组对应的匹配路口组集合。
  5. 根据权利要求4所述的方法,其中,所述根据所述置信值构建匹配的道路网,并计算匹配的道路网的置信值,再通过匹配的道路网构建匹配道路网组合;根据匹配道路网组合中,各道路网的相对位置关系,计算道路网组合匹配的置信值为置信总值,选取置信总值最大的道路网组合为匹配最优解,包括:
    选取所述第一地图数据中的任一路口组JGi为对象,其在所述第二地图数据中匹配的路口组为JG’j;
    执行第一匹配操作,所述第一匹配操作包括:若路口组JGi的相邻路口组JGm在所述第二地图数据中的相匹配的路口组JG’n为路口组JG’j的相邻路口组且置信值大于0,则认为JGi至JGm构成的道路网与JG’j至JG’n构成的道路网相匹配;
    选取路口组JGm为对象,并重复所述第一匹配操作,得到相匹配的路口组JGi和JG’j为中心路口组,JGi和JG’j匹配的所有连通关系构成的道路网Ni和N’j;
    针对以路口组JGi为中心路口组的其中一个道路网Ni,对其相匹配的多个以路口组JG’j为中心路口组的道路网N’j中的各路口组的置信值求和,得到Ni和N’j相匹配的置信值;
    根据路口及其匹配路口不交叉的原则,得到所述第一地图数据中所有 道路网,记为N0至Nn以及所述第二地图数据中对应的道路网组合,记为N’0至N’n,其中n为整数;
    根据匹配关系,在所述第一地图数据和所述第二地图数据中,得到多对N0至Nx的组合,其中x为整数,与n不同或相同;
    根据道路网组合中各道路网的相对位置关系,计算所述第一地图数据中的道路网组合N0至Nn与所述第二地图数据中的道路网组合N’0至N’n匹配的置信值,得到置信总值;
    取置信总值最大的道路网组合为匹配最优解。
  6. 一种基于全局路网特征的数据匹配装置,包括:
    地图加载模块,配置为加载不同地图数据,分别构建各地图中的路口组,并提取路口组属性信息;
    置信值评价模块,配置为根据属性信息的重要度,为路口组的各属性分配权重,综合评价不同地图中的两个路口组相匹配的置信值;
    道路网匹配模块,配置为根据所述置信值构建匹配的道路网,并计算匹配的道路网的置信值,再通过匹配的道路网构建匹配道路网组合;根据匹配道路网组合中,各道路网的相对位置关系,计算道路网组合匹配的置信值为置信总值,选取置信总值最大的道路网组合为匹配最优解。
  7. 根据权利要求6所述的装置,其中,所述地图加载模块包括:
    第一地图加载模块,配置为加载第一地图数据,提取所述第一地图数据中的路口点,并通过道路连通性获取每个路口的相邻路口点;
    第二地图加载模块,配置为加载第二地图数据,提取所述第二地图数据中的路口点,并通过道路连通性获取每个路口的相邻路口点;
    路口组构建模块,配置为通过道路的连通关系分别构建所述第一地图数据和所述第二地图数据的路口组;
    属性提取模块,配置为提取路口组属性信息,所述路口组属性信息包括:路口组之间的连通及位置关系、路口组内路口属性、路口数、组内路 口间的连通及位置关系、组内道路的属性信息以及路口组与路口组之间连通道路的属性信息。
  8. 根据权利要求7所述的装置,其中,所述组内道路的属性信息包括:道路名称、行政区划、道路宽度、道路形状、道路的种别、道路类型。
  9. 根据权利要求7所述的装置,其中,所述置信值评价模块包括:
    权重分配模块,配置为依据经验或深度学习方法,为路口组的各属性分配权重;
    路口组集合生成模块,配置为选取第一地图数据中的任一路口组JGi为对象,在第二地图数据中查找相匹配的路口组JG’j,其中i,j为整数;若在不同的地图数据中,存在一对路口组,具有相同的属性,依据各属性的权重计算这一对路口组匹配的置信值;若没有相同的属性,则匹配的置信值为0;得到第一地图数据中所有路口组对应的匹配路口组集合。
  10. 根据权利要求9所述的装置,其中,所述道路网匹配模块包括:
    匹配道路网生成模块,配置为选取所述第一地图数据中的任一路口组JGi为对象,其在所述第二地图数据中匹配的路口组为JG’j;若路口组JGi的相邻路口组JGm在所述第二地图数据中的相匹配的路口组JG’n为路口组JG’j的相邻路口组且置信值大于0,则认为JGi至JGm构成的道路网与JG’j至JG’n构成的道路网相匹配;
    置信总值计算模块,配置为针对以路口组JGi为中心路口组的其中一个道路网Ni,对其相匹配的多个以路口组JG’j为中心路口组的道路网N’j中的各路口组的置信值求和,得到Ni和N’j匹配的置信值;根据路口及其匹配关系不交叉的原则,得到所述第一地图数据中所有道路网,记为N0至Nn,其中n为整数;根据匹配关系,在所述第一地图数据中,可以得到多个N0至Nx的组合,其中x为整数,与n不同或相同;得到所述第二地图数据中对应的道路网组合,记为N’0至N’m,其中m为整数,且与所述第一地图数据中的道路网个数相同;根据相对位置关系,计算所述第一地图 数据中的道路网组合N0至Nn,与所述第二地图数据中的道路网组合N’0至N’m匹配的置信值,得到置信总值;
    匹配结果输出模块,配置为取置信总值最大的道路网组合为匹配最优解。
  11. 一种基于全局路网特征的数据匹配装置,所述装置包括:
    存储器,配置为保存基于全局路网特征的数据匹配的程序;
    处理器,配置为运行所述程序,其中,所述程序运行时执行权利要求1至5中任一项所述的基于全局路网特征的数据匹配方法。
  12. 一种存储介质,所述存储介质包括存储的程序,其中,所述程序运行时执行权利要求1至5中任一项所述的基于全局路网特征的数据匹配方法。
PCT/CN2019/082132 2018-04-15 2019-04-10 基于全局路网特征的数据匹配方法、装置及存储介质 WO2019201135A1 (zh)

Priority Applications (1)

Application Number Priority Date Filing Date Title
US16/702,719 US11486714B2 (en) 2018-04-15 2019-12-04 Matching algorithm for data with different scales based on global road network features

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201810335065.XA CN108709561B (zh) 2018-04-15 2018-04-15 一种基于全局路网特征的不同尺度数据的匹配算法
CN201810335065.X 2018-04-15

Related Child Applications (1)

Application Number Title Priority Date Filing Date
US16/702,719 Continuation US11486714B2 (en) 2018-04-15 2019-12-04 Matching algorithm for data with different scales based on global road network features

Publications (1)

Publication Number Publication Date
WO2019201135A1 true WO2019201135A1 (zh) 2019-10-24

Family

ID=63866566

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2019/082132 WO2019201135A1 (zh) 2018-04-15 2019-04-10 基于全局路网特征的数据匹配方法、装置及存储介质

Country Status (3)

Country Link
US (1) US11486714B2 (zh)
CN (1) CN108709561B (zh)
WO (1) WO2019201135A1 (zh)

Families Citing this family (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108709561B (zh) 2018-04-15 2020-10-30 武汉中海庭数据技术有限公司 一种基于全局路网特征的不同尺度数据的匹配算法
CN111380537A (zh) * 2018-12-27 2020-07-07 奥迪股份公司 用于在导航地图中定位目标位置的方法及装置
CN112257762B (zh) * 2020-10-12 2022-08-19 武汉中海庭数据技术有限公司 一种异源高精度地图间的路网匹配方法及系统
CN112598305B (zh) * 2020-12-28 2022-02-15 中国人民解放军军事科学院国防科技创新研究院 基于局域结构流量的城市路网重要节点识别方法
CN114323037B (zh) * 2021-12-29 2024-04-12 高德软件有限公司 路段位置匹配、导航方法、装置以及存储介质

Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1787521A (zh) * 2005-12-15 2006-06-14 武汉大学 基于Web地图服务的Web覆盖服务坐标配准方法
JP2007286629A (ja) * 2007-05-11 2007-11-01 Zenrin Co Ltd 電子地図データベースの生成方法
CN101738193A (zh) * 2009-12-29 2010-06-16 北京世纪高通科技有限公司 基于导航电子地图的tmc匹配表的生成方法和装置
CN101788299A (zh) * 2009-12-29 2010-07-28 北京世纪高通科技有限公司 基于导航电子地图的rtic匹配表的更新方法和装置
CN102110364A (zh) * 2009-12-28 2011-06-29 日电(中国)有限公司 基于路口和路段的交通信息处理方法和装置
CN102183259A (zh) * 2011-03-17 2011-09-14 光庭导航数据(武汉)有限公司 基于电子地图道路特征识别的导航方法
CN104850824A (zh) * 2015-04-16 2015-08-19 西安交通大学 一种基于道路元综合的路网匹配方法
CN108709561A (zh) * 2018-04-15 2018-10-26 武汉中海庭数据技术有限公司 一种基于全局路网特征的不同尺度数据的匹配算法

Family Cites Families (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP2597425A1 (en) * 2005-10-14 2013-05-29 Dash Navigation Inc. System and method for identifying road features
CN103852083B (zh) * 2012-12-07 2016-06-29 厦门雅迅网络股份有限公司 一种在导航电子地图数据中增加关联多路口属性的方法
CN104969032A (zh) * 2013-03-15 2015-10-07 惠普发展公司,有限责任合伙企业 地图匹配
US10151592B2 (en) * 2016-04-28 2018-12-11 Here Global B.V. Map matching quality evaluation
CN107292919B (zh) * 2017-06-01 2019-10-18 中国地质大学(武汉) 一种面向空间剖分的多尺度居民地匹配方法

Patent Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1787521A (zh) * 2005-12-15 2006-06-14 武汉大学 基于Web地图服务的Web覆盖服务坐标配准方法
JP2007286629A (ja) * 2007-05-11 2007-11-01 Zenrin Co Ltd 電子地図データベースの生成方法
CN102110364A (zh) * 2009-12-28 2011-06-29 日电(中国)有限公司 基于路口和路段的交通信息处理方法和装置
CN101738193A (zh) * 2009-12-29 2010-06-16 北京世纪高通科技有限公司 基于导航电子地图的tmc匹配表的生成方法和装置
CN101788299A (zh) * 2009-12-29 2010-07-28 北京世纪高通科技有限公司 基于导航电子地图的rtic匹配表的更新方法和装置
CN102183259A (zh) * 2011-03-17 2011-09-14 光庭导航数据(武汉)有限公司 基于电子地图道路特征识别的导航方法
CN104850824A (zh) * 2015-04-16 2015-08-19 西安交通大学 一种基于道路元综合的路网匹配方法
CN108709561A (zh) * 2018-04-15 2018-10-26 武汉中海庭数据技术有限公司 一种基于全局路网特征的不同尺度数据的匹配算法

Also Published As

Publication number Publication date
US20200116500A1 (en) 2020-04-16
CN108709561A (zh) 2018-10-26
CN108709561B (zh) 2020-10-30
US11486714B2 (en) 2022-11-01

Similar Documents

Publication Publication Date Title
WO2019201135A1 (zh) 基于全局路网特征的数据匹配方法、装置及存储介质
JP2019120927A (ja) グリッドマップを作成する方法及び装置
CN105550199B (zh) 一种基于多源地图的点位聚合方法及装置
CN106767836B (zh) 一种auv地形匹配导航滤波方法
CN107292919B (zh) 一种面向空间剖分的多尺度居民地匹配方法
WO2020224305A1 (zh) 用于设备定位的方法、装置及设备
CN109214314B (zh) 一种车道线自动融合匹配算法
CN106777302B (zh) 空间地理坐标的转换方法及装置
US20130311089A1 (en) System and Method for Multi-Plane Routing
CN103712628B (zh) 导航路径描绘方法和终端
CN111080682B (zh) 点云数据的配准方法及装置
JP2015161557A5 (zh)
WO2021232278A1 (zh) 地图获取方法、装置、计算机设备和存储介质
CN105426387A (zh) 一种基于K-means算法的地图聚合方法
Freitas et al. Correcting routing information through GPS data processing
CN111858785B (zh) 地图离散型要素的匹配方法、装置、系统及存储介质
CN112782732A (zh) 一种基于粒子群算法的导航信号解析方法及计算机存可读介质
CN114237265B (zh) 最优日常巡检路线的规划方法、系统、计算机及存储介质
CN111737374B (zh) 位置坐标确定方法、装置、电子设备及存储介质
Abdolmajidi et al. Comparison of matching methods of user generated and authoritative geographic data
Shahidi et al. Graph matching for crowdsourced data in mobile sensor networks
CN112525208A (zh) 城市道路地图快速构建方法和装置及设备
CN111460057A (zh) Poi坐标确定方法、装置与设备
YIN et al. Target Geometry Matching Threshold in Incremental Updatingof Road Networks Based on OSTU
CN110647591A (zh) 用于测试矢量地图的方法和装置

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 19787583

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 19787583

Country of ref document: EP

Kind code of ref document: A1