CN108253976A - It is a kind of fully by the three stage Online Map matching algorithms in vehicle course - Google Patents
It is a kind of fully by the three stage Online Map matching algorithms in vehicle course Download PDFInfo
- Publication number
- CN108253976A CN108253976A CN201810009363.XA CN201810009363A CN108253976A CN 108253976 A CN108253976 A CN 108253976A CN 201810009363 A CN201810009363 A CN 201810009363A CN 108253976 A CN108253976 A CN 108253976A
- Authority
- CN
- China
- Prior art keywords
- point
- gps
- path
- candidate
- vehicle course
- 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.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
- G01C21/00—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
- G01C21/26—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 specially adapted for navigation in a road network
- G01C21/28—Navigation; 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
- G01C21/30—Map- or contour-matching
Abstract
The invention discloses a kind of fully by the Online Map matching algorithm in vehicle course, it is related to map match field, more particularly to the map match carried out by vehicle course.Map match is very important numerous applications based on location information, and to support online wisdom traffic application, online map match is necessary.Therefore, this paper presents a kind of fully by the three stage Online Map matching algorithms in vehicle course, which can not only obtain the high matching result of accuracy, but also with relatively low time complexity.Specifically, first stage, the GPS track point given to one, increase vehicle course information to calculate the GPS point most probable k candidate side in road network；Second stage finds the path between two adjacent GPS track points, reduces search range with vehicle course, and serves as effective search and guide, and accelerates search process；Final stage picks out the true driving path of vehicle with vehicle course information for a series of GPS track points.
Description
Technical field
The present invention relates to map matching technology fields, especially relate to the map-matching algorithm by vehicle course.
Background technology
In recent years, mobile GPS equipment was widely available, and the motion track of user is collected in equipment installation in the car.Vehicle
GPS track be to support the data base based on location information and wisdom traffic service (such as path planning, traffic detect)
Stone.In order to save energy and transmission cost, GPS device records the location information of vehicle with being generally spaced, and which results in track numbers
According to it is openness and uncertain, these reasons can cause reduce apply performance.It is travelled in road network in view of vehicle, in order to
It solves the problems, such as these, GPS track data is mapped on road network, this process is known as map match.Recent years, vehicle are usual
Traffic probe is taken as to detect traffic, in this case, Online Map matching is necessary.
Generally there are two the stage, first stage is looked in the candidate side of each GPS track point for the work of map match
To real matching side, second stage is to infer and find the real driving path of vehicle.The map match work of early stage is logical
Often it is intended merely to find the matching side of GPS track point, but the work can only overcome the measurement error of GPS device.For each
It is candidate when there is the real matching that different probability becomes each given GPS point, pervious work just once utilization orientation information
Information calculates the size of this probability, but directional information used refers to the line direction of former and later two GPS points, this proof
Directional information can help map match, but because next GPS track point can not be learnt during On-line matching, the direction
Information is not used in On-line matching, therefore the present invention is attempted with the vehicle course information being collected into GPS device come so
Complete Online Map matching.
Closer to map match work be primarily upon how reconstructing the driving path of vehicle between two adjacent GPS points.In detail
For thin, the corresponding matching side of two GPS points is unconnected or even apart from far.This work is primarily to solve rail
Mark data it is openness and uncertain.Path is adequately found, such as map-matching algorithm based on Hidden Markov is reasonable
Using many additional informations such as road network topological structure (while with while linking relationship) or road attribute (limiting vehicle speed) increase
The accuracy of matching result is added.In order to be inferred to real driving path, need to find between any pair of adjacent GPS point
Vehicle running path, during this searching path, Dijkstra and A* algorithms are frequently used.But because per a pair of
Path between adjacent GPS point is required for searching out, so finding this process of path extremely expends the time, proposes borrow thus
A kind of didactic path search algorithm in vehicle course is helped, search time can be significantly reduced.
Invention content
In order to ensure the matched high quality of match of Online Map and low time complexity, the present invention provides one fully to borrow
The three stage Online Map matching algorithms in vehicle course are helped, Online Map is carried out to the GPS track data flow being input in algorithm
Matching.For track data stream, it is cut to the original GPS track segment of equal length first, a path segment is exactly one
A processing unit for each path segment data, is matched with three stage match algorithms of proposition.
Specifically, the present invention provides one fully by the tool of the three stage Online Map matching algorithms in vehicle course
Body scheme is：
One fully by the three stage Online Map matching algorithms in vehicle course, including three phases.Wherein, first
Stage is a given GPS track point, determines the candidate side in top-k road network；Second stage is finds two GPS rails
Potential path between mark point；Three phases select the true driving path of vehicle for a series of GPS track points.
Further, the present invention one is fully by the first rank in the three stage Online Map matching algorithms in vehicle course
Section, including following 4 steps：Step 1：To a given GPS track point, first using the GPS point as the center of circle, it is for r meters with length
Radius draws circle (r=100 meter), all when being all the candidate of the GPS point in the road network under the circle covers；Step 2 utilizes
Spatial Probability, calculate that step 1 filters out it is candidate be possibility height when given GPS point really matches；Step 3 utilizes
Direction Probability, calculate that step 1 filters out it is candidate be possibility height when given GPS point really matches；Step 4：According to
Step 2 and the Spatial Probability and Direction Probability on 3 obtained candidate sides calculate the combined chance on every candidate side, to its probability value height
It is low to be ranked up, select the candidate side of top-k maximum probability.
Further, the present invention one is fully by the second-order in the three stage Online Map matching algorithms in vehicle course
Section, including following 2 steps：Step 1 determines potential region of search using the course and travel speed of vehicle；Step 2, root
The tree construction of road network interior joint is established according to vehicle course and road network topology structure, then using deep search algorithm (DFS) latent
Region of search in find path between two GPS track points.
Further, the present invention one is fully by the third rank in the three stage Online Map matching algorithms in vehicle course
Section, is to calculate the probability that every path candidate is coupling path using the topological structure of vehicle course information and road network, then
The path of select probability maximum is as coupling path.
Description of the drawings
Fig. 1 is the schematic diagram of necessary concept in the present invention；
Fig. 2 is triphasic Online Map matching algorithm algorithmic system block diagram；
Fig. 3 is to prove the schematic diagram that vehicle course acts in map match.
Fig. 4 is the path search algorithm by vehicle course.
Specific embodiment
The present invention is further described below in conjunction with the accompanying drawings.
Before algorithm content is introduced, 6 necessary concepts in invention are first stated.
1st concept:Road network, road network are a figure G (N, E), include one group of line set E and a group node set N.It is each
A side eiIt is a directed edge, there are two node, head node n for ithWith tail node nt.Each node n is the seat of a pair of of longitude and latitude
Mark combination represents the spatial position of sampled point.
2nd concept:Side e in road networkiDirection, each edge is there are two direction d (nh,nt),d(nt,nh), d (nh,nt) be
Refer to from node ndTo node ntDirection, this can easily be calculated according to the longitude and latitude of GPS point.Pay attention in institute of the present invention
Direction be all to the north of extremely benchmark.For example, in Fig. 1, d (n5,n6) it is exactly candidate side e8Direction, from node n5
It is directed toward node n6。
3rd concept:Vehicle course, vehicle course h (0 °≤h≤360 °) refer to that headstock of the vehicle in sampling location refers to
To the information can be obtained directly from track data, as shown in Figure 1, vehicle is in GPS sampled points p1Course be h1。
4th concept:GPS track point, GPS track point have recorded the room and time information of vehicle, contain longitude and latitude
Coordinate, timestamp information, instantaneous velocity and vehicle course, are designated as pi=(t1,lati,loni,vi,hi)。
5th concept:Original GPS track segment, original GPS track segment τ, which is one, to be arranged according to time series
GPS track point, be designated as τ=<pi,pi+1,…,pi+l>.Parameter l controls the length of path segment.Original GPS track data
Stream T includes infinite a GPS track segment, is denoted as T=<τ1,τ2,…>。
6th concept:Matched GPS track segment, to an original GPS track segment, its matched GPS track segment
It is vehicle true driving path in road network, is designated as τm=<ei,ei+1,…,en>, the adjacent side of any two of which is all
Connected.One matched GPS track data flow TmComprising infinite matched GPS track segment, it is denoted as Tm=<τm1,τm2,…
>。
One fully by the three stage Online Map matching algorithms in vehicle course, to the GPS track being input in algorithm
Data flow carries out Online Map matching.For track data stream, it is cut to the original GPS track segment of equal length first,
One path segment is exactly a processing unit, for each path segment data, with three stage match algorithms of proposition
It is matched.Algorithm includes three phases, as shown in Fig. 2, first stage is a given GPS track point, determines top-k
Candidate side in a road network；Second stage is the potential path found between two GPS track points；Three phases are a system
Row GPS track point (i.e. GPS track segment) selects the true driving path of vehicle.
1st, first stage includes following four step：
Step 1：Given there are one GPS track points, and first using the GPS point as the center of circle, circle (r=is drawn for radius for r meters with length
100 meters), it is all when being all the candidate of the GPS point in the road network under circle covering.
Step 2：Using Spatial Probability, calculate that step 1 filters out it is candidate be possibility when given GPS point really matches
Property height.Assuming that given GPS track point piWith candidate side ejRange difference beRange difference is typically normal distyribution function,
Expression formula is as follows：Range difference in formulaComputational methods it is as follows:Function dist is for calculating distance between two points, C1
And C2It is candidate side e respectivelyjTwo nodes, C3It is GPS point piTo candidate side ejOn vertical point.Pay attention to if C3Not in side ej
It is upper (i.e. while while ejOutside), then dist (pi,C3)→+∞；σ1It is standard deviation, in order to calculate it, first with existing classics
Map-matching algorithm calculates the matching side of each GPS track point, then solves the matching corresponding with them of all GPS track points
Side apart from difference set, finally easily fit σ1Value.
If the candidate side that GPS track point is found according only to Spatial Probability may result in wrong result.Such as 3 institutes
Show, according to the distance of distance, GPS track point p2It should be with side e1Matching, but what this was clearly wrong.It can be seen that vehicle exists
Sampled point p2Vehicle course closer to e3Direction, this just allow people expect incoming direction probability come for GPS track point screen wait
Select side.
Step 3：Utilization orientation probability, calculate that step 1 filters out it is candidate be possibility when given GPS point really matches
Property height.Assuming that given GPS track point piWith candidate side ejDifferential seat angle beDifferential seat angle is equally normal distyribution function,
Expression formula is as follows：All there are two endpoint n for each edge in road networkhAnd nt, because
This they all there are two direction d (nh,nt) and d (nt,nh), d (nh,nt) direction that refers to side is from endpoint nhTo nt.Angle in formula
The computational methods of difference are as follows：σ2It is also standard deviation, computational methods
With σ1Unanimously.
Step 4：According to Spatial Probability G1With Direction Probability G2, calculate combined chanceTo institute
The combined chance G height for having candidate side is ranked up, and selects the candidate side of top-k maximum probability.
2nd, second stage includes following two steps：
Step 1：Potential region of search is determined using vehicle course and velocity information, it is a sector region, by following
Three conditions uniquely determine：First, the fan-shaped center of circle is first tracing point p in two GPS track pointsi.Second, it is fan-shaped
Radius is equal to WithThe speed of respectively two GPS points, Δ t are two GPS points
Time interval, c are constant (being set as 50 meters).Third, sector are made of two semicircles, and each half diameter of a circle is respectively at two
Vehicle course h on GPS pointiAnd hi+1Vertically.
In potential region of search, the potential path between two adjacent GPS track points is found.Because each GPS point
There is k candidate side, then the adjacent GPS point of any two there may be k in potential region of search2A potential path.It first can root
According to vehicle course information, judge the node in the corresponding two candidate sides of GPS point which be path starting point nsWith terminal ne.Such as
Shown in Fig. 4, a pair of of candidate side e can be easily seen11And e2, node n8And n2It is start node and terminal node respectively.Determine road
After the starting point of diameter, following steps searching route is then utilized.Firstly for all road-net nodes in potential region of search
ni, find its child node nc.Child node ncMeet following two conditions：If ncIt is niChild node, then with niIt compares, ncIt should
It should be from neCloser to, and from nsIt is farther, as shown in formula 3 and 4；And from niTo ncDirection should be with vehicle course hiAnd hi+1
Unanimously, as shown in formula 5.dist(nc,ne)<dist(ni,ne) --- (3), dist (nc,ns)>dist(ni,ns) --- (4), min
(|hi-d(ni,nc)|,|hi+1-d(ni,ns)|)<90 ° --- (5) then according to determining father and son's node relationships, with nsFor root section
Point establishes tree construction using other nodes in potential region of search.If the tree of structure does not include ne, it is somebody's turn to do then can consider
Path is not present between two GPS track points.Otherwise it is found using deep search algorithm (DFS) potential between two tracing points
Path notices that search stops once path is found.
3rd, three phases comprise the steps of：
For being up to k between the adjacent GPS track point of any two2A potential path, then contain l for one
The path segment of a tracing point GPS, at most can there are k2(l-1)The potential route of item.Because each side is corresponding GPS track in path
The probability height that point really matches side is different, so the possibility that every route is the real driving path of vehicle is also different.
The probability size that every paths are the real driving paths of vehicle is calculated with formula (6), the path of final output maximum probability is made
For coupling pathHereRefer to candidate side ejIt is tracing point piVery
The combined chance value on positive match side.Note that the path segment for including l GPS point for one<p1,p2,…,pl>Can energy circuit
Diameter is represented asHereRefer to tracing point piThe candidate side of jth th.It refers to from candidate sideTo candidate sidePath.
Claims (4)
- It is 1. a kind of fully by the three stage Online Map matching algorithms in vehicle course, it is characterised in that include following three stage：(1) in the first stage, it is a given GPS track point, is determined using spatial positional information and vehicle course information Most probable candidate side in top-k road network；(2) in second stage, the potential driving path between two adjacent GPS track points is found using vehicle route information；(3) in the phase III, it is a series of continuous GPS track point sets, picks out the true driving path of vehicle.
- It is 2. according to claim 1 a kind of fully by the three stage Online Map matching algorithms in vehicle course, feature It is that the first stage includes following four step：Step 1：To a given GPS track point, first using the GPS point as the center of circle, drawn with length r for radius for (r=100 meters) Circle, it is all when being all the candidate of the GPS point in the road network under circle covering.Step 2：Using Spatial Probability, calculate that step 1 filters out it is candidate be that possibility when given GPS point really matches is high It is low.Assuming that given GPS track point piWith candidate side ejRange difference beRange difference is typically normal distyribution function, expression Formula is as follows：Range difference in formulaComputational methods it is as follows: Function dist is for calculating distance between two points, C1And C2It is candidate side e respectivelyjTwo nodes, C3It is GPS point piTo time Select side ejOn vertical point.Pay attention to if C3Not in side ejIt is upper (i.e. while while ejOutside), then dist (pi,C3)→+∞；σ1It is Standard deviation, in order to calculate it, we calculate the matching of each GPS track point first with existing classical map-matching algorithm Side, then solve all GPS track points it is corresponding with them matching side apart from difference set, finally easily fit σ1's Value.Step 3：Utilization orientation probability, calculate that step 1 filters out it is candidate be that possibility when given GPS point really matches is high It is low.Assuming that given GPS track point piWith candidate side ejDifferential seat angle beDifferential seat angle is equally normal distyribution function, expression Formula is as follows：All there are two endpoint n for each edge in road networkhAnd nt, therefore they are all there are two direction d (nh,nt) and d (nt,nh), d (nh,nt) direction that refers to side is from endpoint nhTo nt.The computational methods of differential seat angle are as follows in formula： σ2It is also standard deviation, computational methods and σ1Unanimously.Step 4：According to the Spatial Probability G that each all candidate sides of GPS track point are obtained in step 2 and 31With Direction Probability G2, first Calculate the combined chance on each sideThen combined chance G height is ranked up, finally selected Go out the candidate side of top-k maximum probability.
- It is 3. according to claim 1 a kind of fully by the three stage Online Map matching algorithms in vehicle course, feature It is that the second stage includes following two steps：Step 1：Potential region of search is determined using vehicle course and velocity information, it is a sector region, by following three Condition uniquely determines：First, the fan-shaped center of circle is first tracing point p in two GPS track pointsi.Second, fan-shaped radius It is equal to WithThe speed of respectively two GPS points, Δ t are two adjacent GPS points Sampling time interval, c are constant (being set as 50 meters).Third, sector are made of two semicircles, and each half diameter of a circle is respectively with two Vehicle course h on a GPS pointiAnd hi+1Vertically.Step 2：In potential region of search, the potential path between two adjacent GPS track points is found.Because each GPS Point has k candidate side, then the adjacent GPS point of any two may have k in potential region of search2A potential path.It first can be with According to vehicle course information, judge the node in the corresponding two candidate sides of GPS point which be path starting point nsWith terminal ne。 After the starting point for determining path, following steps searching route is then utilized.Firstly for all roads in potential region of search Net node ni, find its child node nc.Child node ncMeet following two conditions：If ncIt is niChild node, then with niPhase Than ncIt should be from neCloser to, and from nsIt is farther, as shown in formula 3 and 4；And from niTo ncDirection should be with vehicle course hiAnd hi+1Unanimously, as shown in formula 5.dist(nc,ne)<dist(ni,ne)---(3)dist(nc,ns)>dist(ni,ns)---(4)min(|hi-d(ni,nc)|,|hi+1-d(ni,ns)|)<90°---(5)Then we are according to determining father and son's node relationships, with nsFor root node, built using other nodes in potential region of search Vertical tree construction.If the tree of structure does not include ne, then it is considered that path is not present between two GPS track points.Otherwise The potential path between two tracing points is found using deep search algorithm (DFS), notices that search stops once path is found.
- It is 4. according to claim 1 a kind of fully by the three stage Online Map matching algorithms in vehicle course, feature It is that the phase III comprises the steps of：For being up to k between the adjacent GPS track point of any two2A potential path, then contain l rail for one The path segment of mark point GPS, at most can there are k2(l-1)The potential route of item.Because each side is that corresponding GPS track point is true in path The probability height on positive match side is different, so the possibility that every route is the real driving path of vehicle is also different.With public affairs Formula (6) calculates the probability size that every paths are the real driving paths of vehicle, the path of final output maximum probability as With path.HereRefer to candidate side ejIt is tracing point piThe really combined chance value on matching side.Note Meaning, the path segment for including l GPS point for one<p1,p2,…,pl>Possible path be represented asHereRefer to tracing point piJ-th candidates side.It refers to From candidate sideTo candidate sidePath.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810009363.XA CN108253976B (en) | 2018-01-04 | 2018-01-04 | Three-stage online map matching algorithm fully relying on vehicle course |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810009363.XA CN108253976B (en) | 2018-01-04 | 2018-01-04 | Three-stage online map matching algorithm fully relying on vehicle course |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108253976A true CN108253976A (en) | 2018-07-06 |
| CN108253976B CN108253976B (en) | 2021-06-15 |
Family
ID=62724887
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810009363.XA Active CN108253976B (en) | 2018-01-04 | 2018-01-04 | Three-stage online map matching algorithm fully relying on vehicle course |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108253976B (en) |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109640263A (en) * | 2018-11-30 | 2019-04-16 | 重庆大学 | A kind of online trace compression system and method based on mobile edge calculations |
| CN110232470A (en) * | 2019-04-11 | 2019-09-13 | 阿里巴巴集团控股有限公司 | A kind of method and device of determining vehicle running path |
| CN110967990A (en) * | 2018-09-30 | 2020-04-07 | 北京地平线信息技术有限公司 | Track determination method and device and electronic equipment |
| CN111857113A (en) * | 2019-04-12 | 2020-10-30 | 北京地平线机器人技术研发有限公司 | Positioning method and positioning device for movable equipment |
| WO2020243937A1 (en) * | 2019-06-04 | 2020-12-10 | Beijing Didi Infinity Technology And Development Co., Ltd. | Systems and methods for map-matching |
| CN112867976A (en) * | 2019-01-23 | 2021-05-28 | 宝马股份公司 | Method for the driverless transfer of a vehicle over a route in an enclosed area |
| CN113286985A (en) * | 2020-09-17 | 2021-08-20 | 华为技术有限公司 | Path planning method and path planning device |
| CN113970333A (en) * | 2021-09-26 | 2022-01-25 | 深圳市跨越新科技有限公司 | Adaptive candidate road searching method, system, terminal device and storage medium |
| CN114674335A (en) * | 2022-03-24 | 2022-06-28 | 西南交通大学 | Optimal target link matching method |
Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20110208426A1 (en) * | 2010-02-25 | 2011-08-25 | Microsoft Corporation | Map-Matching for Low-Sampling-Rate GPS Trajectories |
| CN104331422A (en) * | 2014-10-14 | 2015-02-04 | 广州市香港科大霍英东研究院 | Road section type presumption method |
| CN105095481A (en) * | 2015-08-13 | 2015-11-25 | 浙江工业大学 | Large-scale taxi OD data visual analysis method |
| CN105203116A (en) * | 2015-08-25 | 2015-12-30 | 浙江工业大学 | Map matching method based on conditional random fields and low-sampling-frequency floating car data |
| CN105444769A (en) * | 2015-11-26 | 2016-03-30 | 北京百度网讯科技有限公司 | Map matching method and device |
| CN105489006A (en) * | 2015-12-15 | 2016-04-13 | 浙江工业大学 | Multi-scale road flow visual analysis method based on taxi GPS data |
| CN105526939A (en) * | 2014-09-29 | 2016-04-27 | 高德软件有限公司 | Road coupling method and apparatus thereof |
| CN106023587A (en) * | 2016-05-25 | 2016-10-12 | 电子科技大学 | Track data road network precise matching method based on multi-information fusion |
| CN106157624A (en) * | 2016-08-04 | 2016-11-23 | 浙江工业大学 | Many granularities road shunting visual analysis methods based on traffic location data |
| CN106595680A (en) * | 2016-12-15 | 2017-04-26 | 福州大学 | Vehicle GPS data map matching method based on hidden markov model |
| CN106767873A (en) * | 2016-12-30 | 2017-05-31 | 浙江大学 | A kind of map-matching method based on space-time |
-
2018
- 2018-01-04 CN CN201810009363.XA patent/CN108253976B/en active Active
Patent Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20110208426A1 (en) * | 2010-02-25 | 2011-08-25 | Microsoft Corporation | Map-Matching for Low-Sampling-Rate GPS Trajectories |
| CN105526939A (en) * | 2014-09-29 | 2016-04-27 | 高德软件有限公司 | Road coupling method and apparatus thereof |
| CN104331422A (en) * | 2014-10-14 | 2015-02-04 | 广州市香港科大霍英东研究院 | Road section type presumption method |
| CN105095481A (en) * | 2015-08-13 | 2015-11-25 | 浙江工业大学 | Large-scale taxi OD data visual analysis method |
| CN105203116A (en) * | 2015-08-25 | 2015-12-30 | 浙江工业大学 | Map matching method based on conditional random fields and low-sampling-frequency floating car data |
| CN105444769A (en) * | 2015-11-26 | 2016-03-30 | 北京百度网讯科技有限公司 | Map matching method and device |
| CN105489006A (en) * | 2015-12-15 | 2016-04-13 | 浙江工业大学 | Multi-scale road flow visual analysis method based on taxi GPS data |
| CN106023587A (en) * | 2016-05-25 | 2016-10-12 | 电子科技大学 | Track data road network precise matching method based on multi-information fusion |
| CN106157624A (en) * | 2016-08-04 | 2016-11-23 | 浙江工业大学 | Many granularities road shunting visual analysis methods based on traffic location data |
| CN106595680A (en) * | 2016-12-15 | 2017-04-26 | 福州大学 | Vehicle GPS data map matching method based on hidden markov model |
| CN106767873A (en) * | 2016-12-30 | 2017-05-31 | 浙江大学 | A kind of map-matching method based on space-time |
Non-Patent Citations (2)
| Title |
|---|
| LOU Y，ET AL: "Map-matching for low-sampling-rate GPS trajectories", 《17TH ACM SIGSPATIAL INTERNATIONAL CONFERENCE》 * |
| YUAN J,ET AL.: "An interactive-voting based map matching algorithm", 《2010 ELEVENTH INTERNATIONAL CONFERENCE ON MOBILE DATA MANAGEMENT》 * |
Cited By (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110967990A (en) * | 2018-09-30 | 2020-04-07 | 北京地平线信息技术有限公司 | Track determination method and device and electronic equipment |
| CN110967990B (en) * | 2018-09-30 | 2021-09-03 | 北京地平线信息技术有限公司 | Track determination method and device and electronic equipment |
| CN109640263A (en) * | 2018-11-30 | 2019-04-16 | 重庆大学 | A kind of online trace compression system and method based on mobile edge calculations |
| CN112867976A (en) * | 2019-01-23 | 2021-05-28 | 宝马股份公司 | Method for the driverless transfer of a vehicle over a route in an enclosed area |
| CN110232470B (en) * | 2019-04-11 | 2023-01-31 | 创新先进技术有限公司 | Method and device for determining vehicle driving path |
| CN110232470A (en) * | 2019-04-11 | 2019-09-13 | 阿里巴巴集团控股有限公司 | A kind of method and device of determining vehicle running path |
| CN111857113A (en) * | 2019-04-12 | 2020-10-30 | 北京地平线机器人技术研发有限公司 | Positioning method and positioning device for movable equipment |
| CN111857113B (en) * | 2019-04-12 | 2023-11-03 | 北京地平线机器人技术研发有限公司 | Positioning method and positioning device for movable equipment |
| WO2020243937A1 (en) * | 2019-06-04 | 2020-12-10 | Beijing Didi Infinity Technology And Development Co., Ltd. | Systems and methods for map-matching |
| CN113286985A (en) * | 2020-09-17 | 2021-08-20 | 华为技术有限公司 | Path planning method and path planning device |
| WO2022056770A1 (en) * | 2020-09-17 | 2022-03-24 | 华为技术有限公司 | Path planning method and path planning apparatus |
| CN113970333A (en) * | 2021-09-26 | 2022-01-25 | 深圳市跨越新科技有限公司 | Adaptive candidate road searching method, system, terminal device and storage medium |
| CN114674335A (en) * | 2022-03-24 | 2022-06-28 | 西南交通大学 | Optimal target link matching method |
| CN114674335B (en) * | 2022-03-24 | 2023-06-20 | 西南交通大学 | Matching method of optimal target link |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108253976B (en) | 2021-06-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108253976A (en) | It is a kind of fully by the three stage Online Map matching algorithms in vehicle course | |
| EP3293489B1 (en) | Method and apparatus for providing trajectory bundles for map data analysis | |
| CN110516708B (en) | Path prediction method based on track and road network matching | |
| CN106574975B (en) | Trace matching using peripheral signals | |
| CN107766808A (en) | The method and system that Vehicle Object motion track clusters in road network space | |
| US10352718B2 (en) | Discovering points of entry to a location | |
| EP2556338B1 (en) | Probe data processing | |
| CN104884899B (en) | Determine the method across the track of one or more intersections of transportation network | |
| EP1946044B1 (en) | System and method for identifying road features | |
| Krueger et al. | Visual analysis of movement behavior using web data for context enrichment | |
| US20210055424A1 (en) | Algorithm and architecture for map-matching streaming probe data | |
| CN107228677A (en) | Driftage recognition methods and device | |
| US20120246192A1 (en) | Method of Verifying or deriving Attribute Information of a Digital Transport Network Database Using Interpolation and Probe Traces | |
| CN109459045A (en) | A kind of improvement interactive polls matching process for low frequency GPS track | |
| US11578982B2 (en) | Method and apparatus for map matching trace points to a digital map | |
| Blazquez et al. | Simple map-matching algorithm applied to intelligent winter maintenance vehicle data | |
| Elias et al. | Automatic generation and application of landmarks in navigation data sets | |
| JP5892730B2 (en) | Location history management device, location history management system, server device, location history management method, and program | |
| CN105547310A (en) | Apparatus and method for route planning based on PM2.5 healthy trip | |
| Zhang et al. | Delimited stroke oriented algorithm-working principle and implementation for the matching of road networks | |
| Chen et al. | Local path searching based map matching algorithm for floating car data | |
| CN113295173B (en) | Map matching method for annular road section | |
| Gu et al. | Integration of positioning and activity context information for lifelog in urban city area | |
| Karimi et al. | Uncertainty in personal navigation services | |
| Freitas et al. | Correcting routing information through GPS data processing |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |