CN111694032A - Clustering-based rapid graph matching method for large-scale track data - Google Patents
Clustering-based rapid graph matching method for large-scale track data Download PDFInfo
- Publication number
- CN111694032A CN111694032A CN202010386260.2A CN202010386260A CN111694032A CN 111694032 A CN111694032 A CN 111694032A CN 202010386260 A CN202010386260 A CN 202010386260A CN 111694032 A CN111694032 A CN 111694032A
- Authority
- CN
- China
- Prior art keywords
- track
- clustering
- map matching
- adopting
- road network
- Prior art date
- Legal status (The legal status 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 status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 35
- 230000007704 transition Effects 0.000 claims abstract description 10
- 238000004364 calculation method Methods 0.000 claims abstract description 5
- 238000005259 measurement Methods 0.000 claims description 4
- 230000011218 segmentation Effects 0.000 claims description 3
- 238000006243 chemical reaction Methods 0.000 claims description 2
- 239000011159 matrix material Substances 0.000 claims description 2
- 239000013598 vector Substances 0.000 description 4
- 238000005070 sampling Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 2
- 230000009466 transformation Effects 0.000 description 2
- 230000002159 abnormal effect Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000007418 data mining Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000000717 retained effect Effects 0.000 description 1
- 238000011524 similarity measure Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S19/00—Satellite radio beacon positioning systems; Determining position, velocity or attitude using signals transmitted by such systems
- G01S19/38—Determining a navigation solution using signals transmitted by a satellite radio beacon positioning system
- G01S19/39—Determining a navigation solution using signals transmitted by a satellite radio beacon positioning system the satellite radio beacon positioning system transmitting time-stamped messages, e.g. GPS [Global Positioning System], GLONASS [Global Orbiting Navigation Satellite System] or GALILEO
- G01S19/42—Determining position
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S19/00—Satellite radio beacon positioning systems; Determining position, velocity or attitude using signals transmitted by such systems
- G01S19/38—Determining a navigation solution using signals transmitted by a satellite radio beacon positioning system
- G01S19/39—Determining a navigation solution using signals transmitted by a satellite radio beacon positioning system the satellite radio beacon positioning system transmitting time-stamped messages, e.g. GPS [Global Positioning System], GLONASS [Global Orbiting Navigation Satellite System] or GALILEO
- G01S19/393—Trajectory determination or predictive tracking, e.g. Kalman filtering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
Abstract
The invention discloses a rapid graph matching method of large-scale track data based on clustering. Firstly, dividing an original track by adopting a minimum description length to obtain a track segment set; carrying out density clustering on the track segment set; and extracting a representative track by adopting a scanning method. Secondly, firstly, candidate range query is carried out on the GPS track points to obtain k candidate paths; carrying out probability calculation to obtain observation probability and state transition probability; and rapidly finding the optimal path which maximizes the product of the observation probability and the transition probability in the road network by adopting dynamic planning. The method has the characteristic of high map matching efficiency of large-scale track data, improves the map matching precision to a certain extent, and is a method for improving the map matching efficiency by utilizing track clustering for the first time.
Description
Technical Field
The invention belongs to the technical problems of track clustering and map matching in space-time data mining and intelligent traffic application, and relates to a track clustering algorithm and a map matching algorithm of a GPS track.
Background
With the wide application of positioning devices such as a Global Positioning System (GPS) and the like in an intelligent transportation system, large-scale vehicle driving track data can be easily acquired, and the track data implies a large amount of valuable information. Therefore, the analysis and research of the trajectory data have important significance.
However, due to the limitation of the GPS positioning system, the trajectory data collected by the GPS is often inaccurate and data loss occurs, especially in urban areas where the visibility of the satellites is limited and the satellite signals are easily blocked and refracted by high buildings. Meanwhile, due to the constraint of the power supply and the communication cost of the mobile equipment, most of the acquired tracks are low-frequency sampling data, and the actual driving direction of the vehicle cannot be accurately simulated due to the fact that the sampling time interval of the acquired GPS track data is long. Therefore, in order to determine the driving position of the vehicle on the road, these trajectory data must be preprocessed, i.e. Map Matching (Map Matching), before they can be subsequently analyzed and applied using them. The method mainly aims to match longitude and latitude sampling sequences of the driving track with a digital map road network so as to determine the road on which the vehicle runs.
At present, many map matching algorithms are proposed, and the map matching algorithms can be roughly classified into map matching algorithms based on Kalman filtering, weight, geometry and road network topological structure. These map matching algorithms have the disadvantages of low matching accuracy, poor matching efficiency, and the like. The shortcomings of these map matching algorithms (e.g., based on classical hidden markov and its variants) are more pronounced in the face of large-scale trajectory data that requires map matching. Therefore, a rapid map matching algorithm of large-scale track data based on clustering is provided, the algorithm firstly carries out clustering to extract a representative track, and then carries out map matching on the representative track. To a certain extent, outliers can be filtered out, and the efficiency of map matching is improved. In short, the algorithm has the characteristics of high precision and high efficiency.
Disclosure of Invention
The invention provides a rapid map matching algorithm of large-scale track data based on clustering, aiming at the defects of low matching precision, poor matching efficiency and the like when the existing map matching algorithm faces large-scale track data map matching. The algorithm mainly comprises two steps: firstly, clustering track data and extracting a representative track of the data; secondly, map matching is performed on the representative track. The map matching precision is improved to a certain extent, and the map matching efficiency of large-scale track data is improved.
A rapid map matching algorithm of large-scale track data based on clustering is divided into two stages of track clustering and map matching, and the specific realization of each stage comprises the following steps:
the first stage is as follows: clustering the original track, which comprises the following steps:
the method comprises the following steps: and dividing large-scale vehicle tracks, wherein each track is divided by adopting a Minimum Description Length (MDL) method.
For a set of given tracks, firstly carrying out coordinate conversion, then carrying out track division by adopting an MDL principle, extracting characteristic points of the tracks, carrying out segmentation according to a characteristic point set, and obtaining a track segment set of all vehicles in the same way;
step two: giving defined track distance measurement, and carrying out similarity calculation on the obtained track segment set to obtain a similarity matrix; clustering the track segment set based on the result of the similarity measurement, and adding KD-tree for effective retrieval;
step three: and aiming at the clustering result, extracting the representative track in each cluster by adopting a scanning method to serve as an object for subsequent map matching.
And a second stage: map matching is carried out on the representative track by adopting an HMM map matching algorithm, and the method specifically comprises the following steps:
step four: acquiring road network data and GPS track data, extracting road network information, adding maximum speed information, and establishing an R-tree index for a road network; for each GPS track point, searching all paths in the radius r range in the road network by taking the point as the center of a circle, and determining a group of candidate road sections (k pieces);
step five: calculating observation probability and state transition probability;
step six: and finally, quickly finding the optimal path which maximizes the product of the observation probability and the transition probability in the road network by utilizing Dynamic Programming (DP).
Preferably, the step two clustering method is to use a density-based clustering method-DBSCAN.
The invention has the beneficial effects that:
1. the method can greatly improve the map matching efficiency of the large-scale track data.
2. The method firstly carries out track clustering and filters noise points (abnormal points), thereby improving the map matching precision to a certain extent.
The method has the advantage of efficiently processing the matching of the large-scale track map.
Description of the drawings:
FIG. 1 illustrates the trajectory segmentation step of the present invention;
FIG. 2 is a conceptual diagram of a distance metric according to the present invention;
FIG. 3 illustrates the steps of obtaining a representative trace according to the present invention;
FIG. 4 is a step of trajectory clustering according to the present invention;
FIG. 5 is a schematic diagram of candidate point selection according to the present invention;
FIG. 6 illustrates a map matching procedure according to the present invention;
the specific implementation method comprises the following steps:
the invention is divided into two steps: in the first stage, the tracks are clustered as shown in FIG. 4; the second stage, map matching, is shown in FIG. 6. The method comprises the following concrete steps:
1. track division:
for the obtained trajectory set T ═ T i1,2,.. N }, firstly performing coordinate transformation on all tracks, and then performing coordinate transformation on each track t in the track seti＝{piI | (1, 2.·, n) } is divided, and as shown in fig. 1, a method of minimum description length (MDL principle) is adopted to find the feature points. The MDL principle consists of two parts:
(suppose that
And
is a feature point) and
first, add the first point to the feature point set, let the MDLpar(pi,pj) Is represented by piAnd pjMDL cost of traces in between (i.e., L (H)) + L (D | H)), assuming piAnd pjWhen there is no characteristic point, the original track is retained, so that MDLnopar(pi,pj) Representing the MDL cost. If i is less than k and less than j, MDL is satisfiedpar(pi,pj)≤MDLnopar(pi,pj) Traversing the next point; if MDLpar(pi,pj)＞MDLnopar(pi,pj) The previous point j-1 is inserted into the feature point set. From this point the same procedure was repeated.
2. Density clustering:
and carrying out density clustering on the obtained line segment set. The similarity measure between two line segments is: dist (L)i,Lj)＝w⊥·d⊥(Li,Lj)+w||·d||(Li,Lj)+wθ·dθ(Li,Lj) The three distance values in the formula are obtained from FIG. 2, wherein the vertical distance
Parallel distance d||(Li,Lj)＝MIN(l||1,l||2) Angular distance dθ(Li,Lj)＝||LjI × sin (theta) first step, given two parameters and MinLns, calculating the number of each line segment L in the line segment set in the field range if N isAnd (L) is more than or equal to MinLns, the line segment L is regarded as a core line segment, otherwise, the line segment L is regarded as a noise point. And adding kd-Tree in the step to improve the neighborhood searching efficiency. The second step is that: calculating the density connected set of the core line segments, calculating directlyThe reachable line segments are dense and added to the current cluster. If the newly added line segment is not classified, it is added to the cluster
To do more expansion, since it may be a core line segment; otherwise no addition is required. Given a track cardinality | PTR (C) | for the obtained cluster set, the cluster remains available when the segment cardinality in the resulting cluster is greater than MinLns, otherwise the cluster is deleted.
3. A representative trajectory was obtained:
and extracting a representative track in each cluster by adopting a scanning method, and using the representative track as an object of subsequent map matching as shown in FIG. 3. We first compute the average direction vector, given a set of vectors
The average vector formula is:
| V | is a vector cardinal number; and temporarily rotate the coordinate axes. We then order the start and end points according to the coordinates of the rotation axis. When scanning the start and end points in the sorted order, we count the number of line segments and calculate the average coordinates of these line segments. When the scanning method is used to search the representative track, the relationship between the number of next intersecting line segments and MinLns needs to be judged. If the number of the crossed line segments is larger than MinLns, the rotation is cancelled, and the average coordinate is stored in the representative track coordinate; the final generated set is the coordinate information of the representative trajectory.
4. Candidate point preparation:
and importing road network data and GPS track data, projecting and establishing an R-tree index. For each trajectory T ═ { p > in the representative trajectory set obtained aboveiI 1,2, the diameter r, and selecting all road sections in the road network with the GPS point as the center of a circle and the radius r as candidate road sections
k denotes a GPS point piAnd selecting a point closest to the GPS point on the road section as a candidate point for the kth candidate road section. As shown in fig. 5.
5. Calculating observation probability and state transition probability:
the observation probability is as follows:
piis a GPS track point, xt·iIs a projected point, riIs a projected road section, diThe shortest straight-line distance on a two-point sphere has the inherent meaning if point piDistance riThe closer the "straight line" of the sphere of (a) is, the more we consider p as being from the viewpoint of observationiMore likely at riThe above.
The transition probability is: v (r)i→rj)＝||ri-rj|di-|xt·i-xt+1·jThe intrinsic meaning of route is, assuming two GPS points piAnd pi+1The smaller the deviation between the distance (r) and the route planning distance (route planning distance) between the corresponding projected points on the road, the more likely the vehicle is to be along riTo rjTravel on route.
6. Solving the optimal path:
after the observation probability and the transition probability are obtained, the solution is carried out by using a Viterbi algorithm. With particular reference to the variance in the Gaussian distribution in the observed probability, a true value r may be usediAnd piThe Mean Absolute Deviation (MAD) of the projection distance of (2) is used for parameter estimation. An optimal path that maximizes the product of observation probability and transition probability is quickly found in a road network using Dynamic Programming (DP).
Theoretical proof
The efficiency of the fast map matching algorithm of the large-scale track data based on clustering is the efficiency of the track clustering stage and the map matching stage, and the time complexity of the method is O (N) by simply utilizing the traditional HMM map matching method2). The key point of the invention is to gather highly similar tracks (such as tracks running on the same road) into oneIn each category, when the representative tracks are identified and map matching is carried out, the number of the tracks is greatly reduced from N, and is the same as the number N of the clusters. Due to N>>n, therefore, the efficiency of the fast map matching algorithm based on the clustered large-scale track data is the efficiency of clustering the large-scale track data. The efficiency of the traditional clustering algorithm adopted in the invention is less than O (N)2) The efficiency of the method is improved compared with that of the traditional map matching algorithm. In the clustering process, if a clustering algorithm with lower efficiency is adopted, the improvement of the matching efficiency of the whole map is larger.
Source of innovation points
With the advent of the big data age, trace data grows exponentially. The map matching method has the main idea that each input track is subjected to road network matching with the digital map, and when large-scale track data is processed, the calculation amount of a map matching algorithm is extremely large, so that the traditional map matching algorithm cannot be used for calculating efficiently. When the track data volume is larger, the vehicle track data is denser, and the tracks on the same road section are more, so that similar tracks are gathered together by utilizing track clustering and are divided into a plurality of track clusters, and then a representative track is selected from the clusters to perform map matching. Therefore, the whole track data set is prevented from being calculated, and the calculation amount of map matching is effectively reduced. The map matching precision is improved to a certain extent.
Claims (2)
1. A fast graph matching method of large-scale track data based on clustering is characterized in that: the method comprises two stages of track clustering and map matching, and the specific implementation of each stage comprises the following steps:
the first stage is as follows: clustering the original track, which comprises the following steps:
the method comprises the following steps: dividing large-scale vehicle tracks, wherein each track is divided by adopting a minimum description length method;
for a set of given tracks, firstly carrying out coordinate conversion, then carrying out track division by adopting an MDL principle, extracting characteristic points of the tracks, carrying out segmentation according to a characteristic point set, and obtaining a track segment set of all vehicles in the same way;
step two: giving defined track distance measurement, and carrying out similarity calculation on the obtained track segment set to obtain a similarity matrix; clustering the track segment set based on the result of the similarity measurement, and adding KD-tree for effective retrieval;
step three: aiming at the clustering result, extracting a representative track in each cluster by adopting a scanning method to serve as an object for subsequent map matching;
and a second stage: map matching is carried out on the representative track by adopting an HMM map matching algorithm, and the method specifically comprises the following steps:
step four: acquiring road network data and GPS track data, extracting road network information, adding maximum speed information, and establishing an R-tree index for a road network; for each GPS track point, searching all paths in a radius r range in a road network by taking the point as a circle center, and determining a group of candidate road sections;
step five: calculating observation probability and state transition probability;
step six: and finally, quickly finding the optimal path which maximizes the product of the observation probability and the transition probability in the road network by utilizing dynamic planning.
2. The method for fast graph matching of cluster-based large-scale trajectory data according to claim 1, wherein: the step two clustering method adopts a density-based clustering method-DBSCAN.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010386260.2A CN111694032A (en) | 2020-05-09 | 2020-05-09 | Clustering-based rapid graph matching method for large-scale track data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010386260.2A CN111694032A (en) | 2020-05-09 | 2020-05-09 | Clustering-based rapid graph matching method for large-scale track data |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111694032A true CN111694032A (en) | 2020-09-22 |
Family
ID=72477495
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010386260.2A Pending CN111694032A (en) | 2020-05-09 | 2020-05-09 | Clustering-based rapid graph matching method for large-scale track data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111694032A (en) |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113932821A (en) * | 2021-11-03 | 2022-01-14 | 安徽师范大学 | Track map matching method based on continuous window average direction features |
| CN114353810A (en) * | 2022-01-10 | 2022-04-15 | 河海大学 | HMM efficient map matching method based on R tree and track segmentation |
| CN114485692A (en) * | 2022-03-08 | 2022-05-13 | 安徽师范大学 | High-sampling-rate track data map matching method based on road network connectivity |
| CN114637884A (en) * | 2022-05-16 | 2022-06-17 | 深圳前海中电慧安科技有限公司 | Method, device and equipment for matching cable-stayed cable-computed space-time trajectory with road network |
| CN114879234A (en) * | 2021-10-14 | 2022-08-09 | 电子科技大学 | Important place mining method and device in complex GPS track |
| CN114999151A (en) * | 2022-05-24 | 2022-09-02 | 电子科技大学 | Urban traffic flow hierarchical analysis method and device based on density in GPS track |
| CN116028827A (en) * | 2022-12-31 | 2023-04-28 | 中国电子科技集团公司信息科学研究院 | Track completion method based on vehicle GPS track data clustering |
| CN116578569A (en) * | 2023-07-12 | 2023-08-11 | 成都国恒空间技术工程股份有限公司 | Satellite space-time track data association analysis method |
| EP4300043A1 (en) * | 2022-07-01 | 2024-01-03 | HERE Global B.V. | Method and apparatus for providing a path-based map matcher |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104900059A (en) * | 2015-05-26 | 2015-09-09 | 大连理工大学 | Method for enhancing cell phone base station positioning precision by using Hidden Markov map-matching algorithm |
| CN109459045A (en) * | 2018-09-29 | 2019-03-12 | 杭州电子科技大学 | A kind of improvement interactive polls matching process for low frequency GPS track |
| CN110991475A (en) * | 2019-10-17 | 2020-04-10 | 中国科学院电子学研究所苏州研究院 | Moving object track clustering method based on multi-dimensional distance measurement |
-
2020
- 2020-05-09 CN CN202010386260.2A patent/CN111694032A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104900059A (en) * | 2015-05-26 | 2015-09-09 | 大连理工大学 | Method for enhancing cell phone base station positioning precision by using Hidden Markov map-matching algorithm |
| CN109459045A (en) * | 2018-09-29 | 2019-03-12 | 杭州电子科技大学 | A kind of improvement interactive polls matching process for low frequency GPS track |
| CN110991475A (en) * | 2019-10-17 | 2020-04-10 | 中国科学院电子学研究所苏州研究院 | Moving object track clustering method based on multi-dimensional distance measurement |
Non-Patent Citations (3)
| Title |
|---|
| JAE-GIL LEE ET AL.: ""Trajectory Clustering: A Partition-and-Group Framework"", 《SIGMOD"07:PROCEEDINGS OF THE 2007 ACM SIGMOD INTERNATIONAL CONFERENCE ON MANAGEMENT OF DATA》 * |
| WENTAO BIAN ET AL.: ""A Trajectory Collaboration Based Map Matching Approach for Low-Sampling-Rate GPS Trajectories"", 《SENSORS》 * |
| 严盛隆 等: ""IIVMM：针对低频GPS轨迹的改进交互式投票匹配算法"", 《计算机科学》 * |
Cited By (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114879234A (en) * | 2021-10-14 | 2022-08-09 | 电子科技大学 | Important place mining method and device in complex GPS track |
| CN113932821B (en) * | 2021-11-03 | 2023-06-16 | 安徽师范大学 | Track map matching method based on continuous window average direction characteristics |
| CN113932821A (en) * | 2021-11-03 | 2022-01-14 | 安徽师范大学 | Track map matching method based on continuous window average direction features |
| CN114353810A (en) * | 2022-01-10 | 2022-04-15 | 河海大学 | HMM efficient map matching method based on R tree and track segmentation |
| CN114485692A (en) * | 2022-03-08 | 2022-05-13 | 安徽师范大学 | High-sampling-rate track data map matching method based on road network connectivity |
| CN114485692B (en) * | 2022-03-08 | 2023-06-16 | 安徽师范大学 | High sampling rate track data map matching method based on road network connectivity |
| CN114637884A (en) * | 2022-05-16 | 2022-06-17 | 深圳前海中电慧安科技有限公司 | Method, device and equipment for matching cable-stayed cable-computed space-time trajectory with road network |
| CN114637884B (en) * | 2022-05-16 | 2022-08-23 | 深圳前海中电慧安科技有限公司 | Method, device and equipment for matching cable-stayed cable-computed space-time trajectory with road network |
| CN114999151B (en) * | 2022-05-24 | 2023-01-24 | 电子科技大学 | Density-based urban traffic flow hierarchical analysis method and device in GPS track |
| CN114999151A (en) * | 2022-05-24 | 2022-09-02 | 电子科技大学 | Urban traffic flow hierarchical analysis method and device based on density in GPS track |
| EP4300043A1 (en) * | 2022-07-01 | 2024-01-03 | HERE Global B.V. | Method and apparatus for providing a path-based map matcher |
| CN116028827A (en) * | 2022-12-31 | 2023-04-28 | 中国电子科技集团公司信息科学研究院 | Track completion method based on vehicle GPS track data clustering |
| CN116578569A (en) * | 2023-07-12 | 2023-08-11 | 成都国恒空间技术工程股份有限公司 | Satellite space-time track data association analysis method |
| CN116578569B (en) * | 2023-07-12 | 2023-09-12 | 成都国恒空间技术工程股份有限公司 | Satellite space-time track data association analysis method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111694032A (en) | Clustering-based rapid graph matching method for large-scale track data | |
| CN110095127B (en) | Hidden Markov model map matching method based on segmentation | |
| CN112015835B (en) | Geohash compressed map matching method | |
| CN110909788B (en) | Statistical clustering-based road intersection position identification method in track data | |
| CN108253976B (en) | Three-stage online map matching algorithm fully relying on vehicle course | |
| CN103149576A (en) | Map matching method of floating car data | |
| CN107917716B (en) | Fixed line navigation method, device, terminal and computer readable storage medium | |
| Qiu et al. | Automatic extraction of road networks from GPS traces | |
| CN113932821B (en) | Track map matching method based on continuous window average direction characteristics | |
| CN112504265A (en) | Geomagnetic reference library construction method for indoor vehicle geomagnetic matching positioning | |
| CN112991722B (en) | High-frequency gps (gps) point bus real-time intersection prediction method and system | |
| CN109307513B (en) | Real-time road matching method and system based on driving record | |
| CN113959452A (en) | Map matching method, system and terminal based on urban road network | |
| CN113295173A (en) | Map matching method for annular road section | |
| CN112309118A (en) | Vehicle trajectory calculation method based on space-time similarity | |
| CN114935773A (en) | Online road matching method, device, medium and equipment based on Beidou positioning | |
| CN113611115B (en) | Vehicle track clustering method based on road network sensitive characteristics | |
| Yu et al. | High-frequency trajectory map matching algorithm based on road network topology | |
| CN114993325A (en) | Vehicle GPS map road matching method, device and storage medium | |
| CN114139099A (en) | Road intersection information extraction method based on track density homogenization and hierarchical segmentation | |
| CN114353810B (en) | HMM efficient map matching method based on R tree and track segmentation | |
| CN110727749A (en) | Self-adaptive candidate road section searching method | |
| CN114061597B (en) | Autonomous positioning method for vehicle map matching | |
| CN116828397B (en) | Track information acquisition method and device, electronic equipment and storage medium | |
| CN112013842A (en) | Multi-mode indoor positioning method based on image geomagnetic field and inertial sensor |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200922 |