CN105628033B - A kind of map-matching method based on path connected relationship - Google Patents
A kind of map-matching method based on path connected relationship Download PDFInfo
- Publication number
- CN105628033B CN105628033B CN201610107068.9A CN201610107068A CN105628033B CN 105628033 B CN105628033 B CN 105628033B CN 201610107068 A CN201610107068 A CN 201610107068A CN 105628033 B CN105628033 B CN 105628033B
- Authority
- CN
- China
- Prior art keywords
- gps
- section
- road
- point
- matching
- 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.)
- Active
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
Landscapes
- Engineering & Computer Science (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Automation & Control Theory (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Navigation (AREA)
- Position Fixing By Use Of Radio Waves (AREA)
Abstract
The present invention provides a kind of map-matching method based on path connected relationship, comprising the following steps: step S1 pre-processes GPS positioning data；Step S2, path adaptation rule is established according to distance and two, direction factor, and then candidate matches road set is obtained, the similitude between GPS track and candidate matches road is calculated based on path connected, and then select matching road of the similitude the maximum as GPS track；Step S3 carries out coordinate adjustment to GPS track.The wrong data that the present invention is directed to different type of errors respectively proposes corresponding detection and processing method, ensure that the high efficiency and accuracy of map-matching method；Not only it had considered the correlation of historical data, but also has reduced influence of the biggish point of discrete error to direction calculating；The binding character of history match information is also contemplated, therefore matching result accuracy rate is higher, overall technical architecture is more simpler than the model of the prior art, and continuity and matching effect are more preferable.
Description
Technical field
The present invention relates to a kind of map-matching method more particularly to a kind of map match sides based on path connected relationship
Method.
Background technique
In recent years, Satellite Navigation Technique, especially GPS (Global Positioning System,
GPS it) is widely applied in terms of road traffic positioning, utilization of the GPS in terms of traffic positioning includes path navigation, vehicle
Scheduling, accident emergency reaction, location tracking and traffic information acquisition etc..As these are based on location technology location-based service
The fast development of (Location Based Service, LBS), a large amount of track of vehicle data constantly generate will be to these tracks
Data are analyzed, and then excavating useful information, preferably providing information service is when previous research hotspot.
However, often there are one with actual geographic position for GPS positioning result since GPS positioning data often have error
Fixed to deviate, this causes difficulty to location based service, therefore, it is necessary on the basis of the relatively high electronic map of precision,
On the road that GPS positioning path matching to vehicle is really run, here it is map match.
By the analysis to data with existing, GPS positioning error information is mainly shown as following three form: one, data are superfluous
Remaining, when vehicle is when a certain position is static or low speed is run, vehicle can record a series of identical GPS of coordinates in same place
The location information of point, these redundancies will affect the effect of map match, even will appear error situation sometimes；Two, shortage of data,
When there is bad signal shielding, signal, cold start-up, instrument failure, power depletion, maloperation, one may be generated
The shortage of data of a period, shortage of data cause the discontinuous of GPS positioning track, can equally reduce map match effect；Three,
Data wander, near the starting point of driving path, terminal or other, it may occur that GPS positioning data, which are not met, patrols
It collects, serious the phenomenon that deviating vehicle physical location, i.e. data wander.Drift data is a kind of noise, and meeting to map is matched and shown
Show that effect makes a big impact.
For above-mentioned GPS data error condition, need to find corresponding processing method, and to the GPS track after processing
Map match is carried out, by GPS track data correction to physical location, to meet the needs of location information service.
Currently, the research achievement about map match is existing very much；There is fairly simple matching process, as closest approach matches
Method；Also there is complicated matching process, such as fuzzy logic algorithm, the method based on cost function, the method based on D-S evidential reasoning
And method neural network based etc.；Closest approach matching process judged at a distance from road according to anchor point, this method
Simply, easy to accomplish, but matching effect is poor；The main think of of weight factor algorithm, fuzzy logic algorithm and algorithm for pattern recognition
Think to be searching and matching result of the similitude maximum road in GPS positioning track as the track in electronic map road net,
Although the effect of these methods has a distinct increment compared with closest approach matching process, model is complex, and system-computed is opened
Sell larger.
Summary of the invention
Relatively simple the technical problem to be solved by the present invention is to need to provide a kind of model, continuity is good and matching effect
Better map-matching method.
In this regard, the present invention provides a kind of map-matching method based on path connected relationship, comprising the following steps:
Step S1 pre-processes GPS positioning data；
Step S2 establishes path adaptation rule according to distance and two, direction factor, and then obtains candidate matches road collection
It closes, the similitude between GPS track and candidate matches road is calculated based on path connected, and then similitude the maximum is selected to make
For the matching road of GPS track；
Step S3 carries out coordinate adjustment to GPS track.
A further improvement of the present invention is that in the step S1, for data redundancy, shortage of data and data wander point
It is not removed or interpolation operation, realizes the pretreatment to GPS positioning data.
A further improvement of the present invention is that in the step S1, when GPS positioning data are there are when polyisomenism, will weigh
Multiple GPS positioning data are deleted；When vehicle stops and generates the identical GPS positioning data of two or more position coordinates, then protect
It stays the GPS positioning data at nearest time point and records the vehicle dwell time；It is fixed when lacking GPS between two adjacent GPS points
When the data of position, interpolation processing is carried out；When detecting drift data, delete processing is carried out to drift data.
A further improvement of the present invention is that if time difference Δ t between two adjacent GPS pointspIt is full with sampling interval Δ t
Sufficient formula Δ tp=ti-ti+1≠ Δ t, then it represents that there are missing datas between the two adjacent GPS points, and then according to formulaCarry out interpolation processing；GPS point A, B and C continuous for three are constituted
Triangle Δ ABC, the area s of triangle Δ ABC isWherein, p is three
The half of angular Δ ABC perimeter, a, b and c are respectively three side lengths of triangle Δ ABC, if GPS point is located at triangle Δ ABC
B point, and B point between its adjacent front and back two o'clock A point and C point line distance meetWherein,
derrorWhat is indicated is the maximum value for allowing offset, and it is that the equipment positioning accuracy of GPS point and road are had a lot of social connections that this, which allows the maximum value of offset,
The sum of degree then deletes the GPS positioning data of the point at this point, showing that B point is shift point.
A further improvement of the present invention is that establishing road according to distance and two, direction factor in the step S2
It is the building that topological relation and section connected relation are carried out according to two factors of distance and direction with rule, and by the section of building
Topological relation between connected relation and dotted line is saved in serializing file.
A further improvement of the present invention is that the step S2 further includes following sub-step:
Step S201 calculates error band by the distribution of the GPS positioning probability of error and road width factor, will fall in error
Gather as the candidate matches section of the GPS point in all sections in region；
Step S202, by the direction difference between the distance between GPS point and section and GPS track and section to time
Choosing matching section scoring, the comprehensive score in candidate matches section are apart from the sum of score and direction score；
Step S203, by the matching result and path connected relationship of a upper GPS point, from the candidate matches road scored
Optimum Matching road is selected in Duan Jihe.
A further improvement of the present invention is that calculating error band by formula R=r+w, in turn in the step S201
Obtain candidate matches road set of all sections in the border circular areas that radius is R as the GPS point, wherein w has a lot of social connections for road
Degree,R is that the radius of candidate matches road set calculates
Value, c ' and d respectively indicate the axial length of error ellipse,WithFor the standard value of position error,For the initial of position error
Value；And the parameters of error ellipse are as follows:
WithWherein, θ indicates the angle of error ellipse long axis and y-axis,Indicate unit weight
Posteriori error,Indicate covariance；δxAnd δyFor the position error of unit weight, δxyFor the covariance of unit weight.
A further improvement of the present invention is that in the step S202, if the upright projection point Q' of GPS point Q to section AB
It falls on the section, then the GPS point is the distance between Q and Q' d (Q, Q ') at a distance from section；It otherwise, is GPS point to the road
The smaller of section two-end-point distance；Pass through Ws′It indicates apart from score, then GPS point piWith target road section ejApart from score are as follows:Wherein μ=0, σ are equal to candidate roads zone radius R；Its direction score passes through formula Wa=cos α
It is available, wherein note v is direction of vehicle movement, and g is road bearing mark, and g=0 indicates two-way traffic road, g=1
Indicate that, along GPS point number order one-way trip, g=2 is indicated along GPS point number order backward one-way trip, piAnd pi+1Respectively
Section to be matched is along the rear and front end point in GPS point number direction, then vehicle traffic direction and road direction angle α are as follows:S ' is distance.
A further improvement of the present invention is that passing through formula W=k in the step S203s′·Ws′+kα·WaIt calculates and waits
The comprehensive score in choosing matching section, wherein ks′For distance weighting, kαFor direction weight,
A further improvement of the present invention is that carrying out the realization process of coordinate adjustment to GPS track in the step S3
Are as follows: when the subpoint of the fitting a straight line endpoint of GPS track is fallen on candidate matches section, then the projection of the fitting a straight line endpoint it
Between distance be part to be matched；If the subpoint of fitting a straight line endpoint is fallen in except candidate matches section, candidate should be selected
The corresponding end point in matching section is matched, what coordinate adjustment process obtained after the geometric transformation of translation, rotation and scaling
Coordinate is final matching results.
Compared with prior art, the beneficial effects of the present invention are: summarize data redundancy, data in GPS positioning data
The type of error of missing and data wander, and corresponding detection and processing are proposed for the wrong data of different type of errors respectively
Method ensure that the high efficiency and accuracy of the map-matching method；And minimum is used when obtaining direction of vehicle movement
Two multiply fitting a straight line method, not only consider the correlation of historical data, but also reduce the biggish point of discrete error to direction calculating
Influence；On this basis, it is also contemplated that the binding character of history match information, therefore matching result accuracy rate is higher, whole skill
Art scheme is more simpler than the model of the prior art, and continuity and matching effect are more preferable.
Detailed description of the invention
Fig. 1 is the workflow schematic diagram of an embodiment of the present invention；
The detailed operation flow diagram of Fig. 2 an embodiment of the present invention；
Fig. 3 is the recognition methods schematic diagram of the GPS drift data of an embodiment of the present invention；
Fig. 4 is the path connected relation schematic diagram of an embodiment of the present invention；
Fig. 5 is the error ellipse schematic diagram of an embodiment of the present invention；
Fig. 6 is the schematic diagram of the acquisition candidate matches road set of an embodiment of the present invention；
Fig. 7 is the least square method fitting a straight line schematic diagram of an embodiment of the present invention；
Fig. 8 is the calculating section transition probability process schematic of an embodiment of the present invention；
Fig. 9 is the result schematic diagram of the calculating section transition probability of an embodiment of the present invention；
Figure 10 is the coordinate adjustment schematic diagram of an embodiment of the present invention；
Figure 11 is emulation when an embodiment of the present invention chooses simple path by matching obtained Optimum Matching road
Figure；
Figure 12 is imitative after GPS track to be corrected to Optimum Matching road when an embodiment of the present invention chooses simple path
True result figure；
Figure 13 is emulation when an embodiment of the present invention chooses complicated road by matching obtained Optimum Matching road
Figure；
Figure 14 is imitative after GPS track to be corrected to Optimum Matching road when an embodiment of the present invention chooses complicated road
True result figure；
The experimental result analogous diagram of Figure 15 geometric similarity matching process in the prior art；
Figure 16 is the experimental result analogous diagram of the map match of an embodiment of the present invention.
Specific embodiment
With reference to the accompanying drawing, preferably embodiment of the invention is described in further detail.
As depicted in figs. 1 and 2, this example provides a kind of map-matching method based on path connected relationship, including following step
It is rapid:
Step S1 pre-processes GPS positioning data；
Step S2 establishes path adaptation rule according to distance and two, direction factor, and then obtains candidate matches road collection
It closes, the similitude between GPS track and candidate matches road is calculated based on path connected, and then similitude the maximum is selected to make
For the matching road of GPS track；
Step S3 carries out coordinate adjustment to GPS track, for GPS track to be corrected to corresponding according to dependency rule
With on road.
In step S1 described in this example, it is removed respectively for data redundancy, shortage of data and data wander or interpolation is grasped
Make, realizes the pretreatment to GPS positioning data；In the step S1, when GPS positioning data are there are when polyisomenism, it will repeat
GPS positioning data delete；When vehicle stops and generates the identical GPS positioning data of two or more position coordinates, then retain
The GPS positioning data at nearest time point simultaneously record the vehicle dwell time；When lacking GPS positioning between two adjacent GPS points
When data, interpolation processing is carried out；When detecting drift data, delete processing is carried out to drift data.
The step S1 pre-processes the GPS positioning data of GPS track, is put in storage tool for original positioning by GPS
Data are imported in database and are constructed in the database and index, to improve efficiency data query.
If the time difference Δ t between two adjacent GPS pointspMeet formula Δ t with sampling interval Δ tp=ti-ti+1≠ Δ t,
Then indicate that there are missing datas between the two adjacent GPS points, and then according to formula Carry out interpolation processing.
As shown in figure 3, the triangle Δ ABC that GPS point A, B and C continuous for three are constituted, the face of triangle Δ ABC
Accumulating s isWherein, p is the half of triangle Δ ABC perimeter, and a, b and c are respectively
Three side lengths of triangle Δ ABC, if GPS point is located at the B point of triangle Δ ABC, and B point is to its adjacent front and back two o'clock A point and C
The distance of line meets between pointWherein, derrorWhat is indicated is the maximum value for allowing offset, this is allowed
The maximum value of offset is the sum of equipment positioning accuracy and road width of GPS point, at this point, showing that B point is shift point, then by the point
GPS positioning data delete.
In step S2 described in this example, path adaptation rule is established as according to distance and side according to distance and two, direction factor
The building of topological relation and section connected relation is carried out to two factors, and will be between the section connected relation and dotted line of building
Topological relation is saved in serializing file.
Step S2 described in this example carries out topological relation and the connected relation building of road network, and formation sequence file；Root
According to the connection relationship of road network intermediate node and line, dotted line topological relation is constructed；As shown in figure 4, being closed according to the connection between section
System constructs 1,2,3 grade of run-through large space, and it is 0 cascade by extension road itself that connectivity, which extends rank using road to be measured,
Logical collection C0, extends the available 1 grade of connected set C1 of level-one, continue to extend available 2 grades of connected set C2 ..., n grades of connected sets
Cn.By analyzing GPS data, 3 grades of run-through large spaces can meet the matched requirement of map between constructing section.By the section of building
Topological relation between connected relation and dotted line is saved in serializing file, the topological relation in this way between needing to use road network
When can directly read serializing file, avoid rebuilding topological relation and connected relation every time, to improve matching efficiency.
It further includes following sub-step that step S2 described in this example, which choose to road:
Step S201 calculates error band by the distribution of the GPS positioning probability of error and road width factor, will fall in error
Gather as the candidate matches section of the GPS point in all sections in region；
Step S202, by the direction difference between the distance between GPS point and section and GPS track and section to time
Choosing matching section scoring, the comprehensive score in candidate matches section are apart from the sum of score and direction score；
Step S203, by the matching result and path connected relationship of a upper GPS point, from the candidate matches road scored
Optimum Matching road is selected in Duan Jihe.
For step S201 described in this example for obtaining candidate matches road set, road network is large number of, it is impossible to will own
Section is as candidate road section, it is therefore desirable to lay down a regulation and filter out possible candidate roads set.Since GPS positioning error meets
Probability statistics rule, location data centered on physical location, are distributed in an error ellipse region always, as shown in figure 5,
According to Probability Statistics Theory, error ellipse parameter are as follows:
WithWherein, c ' and d respectively indicates the axial length of error ellipse,WithFor the mark of position error
Quasi- value, θ indicate the angle of error ellipse long axis and y-axis,Indicate the posteriori error of unit weight,Indicate covariance.
Assuming that space two-dimensional the coordinate x and y that positioning system obtains are mutually indepedent, and the direction x is identical with the variance in the direction y,
Then elliptic region becomes round region, and radius formula is calculated by following equation: Furthermore, it is contemplated that the influence of road width factor, candidate matches road area radius R value is R=r+
W, wherein w is road width, and r is the radius calculated value of candidate matches road set,For the initial value of position error, this is fixed
The initial value of position error can make the difference value by actual measurement and verifying and obtain according to different positioning device different froms.
Above-mentioned candidate matches road area radius R value has comprehensively considered the distribution of the GPS positioning probability of error and road width
Factor is fallen in centered on GPS point, and R is the candidate road section set that all sections in the border circular areas of radius are the GPS point；
The quantity that candidate road section can be greatly reduced in this way, improves matching efficiency.As shown in fig. 6, section l1、l2、l3It falls in GPS
Point pi be the center of circle, R be radius circle in, therefore, l1、l2、l3For piCandidate matches road.
Step S202 described in this example is used to score to candidate matches road, since GPS point is smaller at a distance from section, GPS
A possibility that angle is smaller between track and section, then the section is GPS track matching section is bigger, therefore can pass through GPS point
It scores with the direction difference at a distance from section and between GPS track and section to candidate matches road.
Have at a distance from section for GPS point defined below: if the upright projection point Q' of GPS point Q to section AB falls in this
On section, then the GPS point is the distance between Q and Q' d (Q, Q ') at a distance from section；It otherwise, is GPS point to the section both ends
The smaller of point distance.
Pass throughIt is found that being analyzed according to data, GPS point and target
Relationship between a possibility that distance in section matches object with the section for GPS point meets the probability density letter of normal distribution
Number, therefore, can be indicated apart from score with the probability density function of normal distribution.
Pass through Ws′It indicates apart from score, then GPS point piWith target road section ejApart from score are as follows: Wherein μ=0, σ are equal to candidate roads zone radius R.
Direction score described in this example passes through formula Wa=cos α is available, wherein note v is direction of vehicle movement, and g is
Road bearing mark, g=0 indicate that two-way traffic road, g=1 are indicated along GPS point number order one-way trip, g=2 table
Show along GPS point number order backward one-way trip, piAnd pi+1Front and back two of the section respectively to be matched along GPS point number direction
Endpoint, then vehicle traffic direction and road direction angle α are as follows:
S ' is distance.
Most map-matching methods are all using GPS point course data or by the direction of consecutive points line as vehicle movement side
To.However, single GPS point course may be influenced by situations such as inertia or sideslip；Adjacent GPS point line is only only in accordance with two
Point determines that the direction of motion, accuracy depend on location data precision.In view of the foregoing, the present invention is straight using least square fitting
Collimation method obtains direction of vehicle movement, as shown in fig. 7, the direction of the straight line obtained using linear least squares fit is as vehicle fortune
Dynamic direction.This method both with respect to the correlation of historical data, in turn avoided the more a little bigger influence of discrete error.
As shown in fig. 7, so-called least square method fitting a straight line, i.e., by a series of measurement point p1(x1,y1)、p2(x2,
y2)、…、pn(xn,yn) track replaced with straight line y=ax+b approximation.Wherein the calculation formula of a, b are as follows:Straight line y=ax+b is used as vehicle movement along the direction of GPS point number order
Direction.
A possibility that direction of vehicle movement and candidate roads direction are closer, matching is then bigger, i.e. vehicle heading
It is bigger with the smaller then corresponding direction fractional value of road direction angle.When angular range arrives 180 degree for 0, cosine function property
Just the characteristics of meeting direction score.When there are multiple sections in conversion section or turning at intersection, due to vehicle
Driving direction is larger to the matching constraint of GPS point, thus the weight that accounts for of direction score also answer it is larger.
In step S203 described in this example, pass through formula W=ks′·Ws′+kα·WaCalculate the comprehensive of candidate matches section
Point, wherein ks′For distance weighting, kαFor direction weight,
If current GPS point is in simple straight road, distance weighting kdIt is larger；If GPS point is in intersection
In mouth, turning or other more complex road networks, then D-factor is larger to the matching constraint of GPS point, direction weight kαIt answers
It is larger.
Since in section intersection, turning or other Complicated Road Networks, generally there are three kinds of operating statuses for vehicle:
Parking waiting, straight trip pass through, turn round.The speed of vehicle is smaller in the case of three kinds at this time, thus can by judge velocity amplitude into
Row weight coefficient assignment (threshold speed is set as 5m/s herein):It is calculated accordingly
Comprehensive score it is higher, the candidate roads be GPS track matching object a possibility that it is bigger.
Step S203 described in this example is for realizing path selection.Consider the matching result and path connected of a upper GPS point
Sexual factor selects Optimum Matching road from the candidate matches road concentration to have scored.As shown in Figure 8 and Figure 9, specific calculating side
Method is as follows.
The first, section transition probability between path connected set；Since vehicle running track has the characteristics that space-time expending,
So matching section corresponding to adjacent GPS point should have connected relation, therefore the conversion between adjacent GPS point candidate roads is general
Rate can be measured with the topology connectivity between road.Road extension rank can be used to be measured in this connected relation.
The floating car data sampling time interval that this example uses is 40s, most of adjacent according to the analysis to experimental data
GPS point is in same or level-one communication channel road, is partially present in secondary road, less greater than three-level connection road.Cause
Transition probability between this adjacent segments sets as follows: present road is connected to then probability with next road zero level, and as 1, level-one connection is general
Rate 0.9, second level connected probability 0.8, being more than or equal to three-level connection, then probability is 0.4；Transition probability can table between path connected set
It is shown as:
The second, Optimum Matching path is obtained based on comprehensive score and candidate road section transition probability, enabled For GPS point piCandidate roads set,For candidate roadsComprehensive score,Indicate pi-1The
J candidate roads are to piKth candidate roads transition probability, vehicle is obtained based on comprehensive score and candidate roads transition probability
The Optimum Matching path of track.
If GPS point piCorresponding Optimum Matching pathFinal score beAs i=1,That is road
The initial final score of diameter is section itself comprehensive score), otherwise the final score in current candidate path is corresponding for a upper GPS point
Optimum Matching path to the transition probability in current candidate path and the product of current candidate path comprehensive score, i.e.,
All candidate matches path final score the maximum are current GPS point piOptimum Matching path be
As shown in Figure 10, in step S3 described in this example, the realization process of coordinate adjustment is carried out to GPS track are as follows: when GPS rail
The subpoint of the fitting a straight line endpoint of mark is fallen on candidate matches section, then the distance between the projection of the fitting a straight line endpoint is
Part to be matched；If the subpoint of fitting a straight line endpoint is fallen in except candidate matches section, candidate matches section should be selected
Corresponding end point is matched, and the coordinate that coordinate adjustment process obtains after the geometric transformation of translation, rotation and scaling is most
Whole matching result.
In the step 3, coordinate adjustment is carried out to GPS point, each GPS point obtains unique matching section, by GPS point to
It is projected on the section, restitution point coordinate of the GPS point in corresponding road section can be obtained.Specific step is as follows:
Step S301, if GPS track point p1,p2,…,pnFitting a straight line AB terminal A, B subpoint a, b fall in matching
On the l of section, then line segment ab is part to be matched.
Step S302 should select the corresponding end point of l to be matched if a or b are fallen in except l.
If the original coordinates matrix of GPS track point set P isCoordinate adjustment process is by translation
T1, rotation T2 and scaling T3 geometric transformation three times, transformation matrix difference are as follows:With
Coordinates matrix after then GPS track point is correctedP ' is final
With result.
This example summarizes the type of error of data redundancy in GPS positioning data, shortage of data and data wander, and needle respectively
Corresponding detection and processing method are proposed to the wrong data of different type of errors, to guarantee the efficient of the map-matching method
Property and accuracy provide precondition；And least square fitting linear method is used when obtaining direction of vehicle movement, both
In view of the correlation of historical data, and reduce influence of the biggish point of discrete error to direction calculating；On this basis, also
The binding character of history match information is considered, therefore matching result accuracy rate is higher, mould of the overall technical architecture than the prior art
Type is more simple, and continuity and matching effect are more preferable.
Finally, being using analysis of experimental results of the invention: based on the above map-matching algorithm step, this example gives two
Group is tested and is analyzed result.
First group is to choose simple road conditions to carry out matching experiment, is for the map match side according to this example shown in Figure 11
The optimal road that method matches is that GPS track is corrected to the matching by map-matching method described in this example shown in Figure 12
Result after road.
The more complex road conditions of second group of selection carry out matching experiment, are the map match sides according to this example shown in Figure 13
The optimal road that method matches is that GPS track is corrected to the matching by map-matching method described in this example shown in Figure 14
Result after road.
The map-matching method described in this example it can be seen from test result is equal to simple path situation and complicated road conditions
There is preferable matching effect, GPS track point can be corrected on correct road.In addition, being respectively shown in Figure 15 and Figure 16
According to the office of the matching process experimental result of traditional geometrical factor similitude and the mentioned map-matching method experimental result of this example
Portion's comparison diagram.It as shown in figure 15, is conventional method experimental result, since traditional geometric match method is only according to distance, side
To etc. factors similitude the maximum is chosen from candidate matches road set as final matching object, without considering path connected
Sexual factor, therefore will appear the illogical situation of matching result.The map-matching method reality mentioned as shown in figure 16 by this example
It tests as a result, since context of methods considers path connected sexual factor, joined in the matching process in same link
The constraint factor of road transition probability, therefore matching result is more accurate.
The above content is a further detailed description of the present invention in conjunction with specific preferred embodiments, and it cannot be said that
Specific implementation of the invention is only limited to these instructions.For those of ordinary skill in the art to which the present invention belongs, exist
Under the premise of not departing from present inventive concept, a number of simple deductions or replacements can also be made, all shall be regarded as belonging to of the invention
Protection scope.
Claims (6)
1. a kind of map-matching method based on path connected relationship, which comprises the following steps:
Step S1 pre-processes GPS positioning data；
Step S2 establishes path adaptation rule according to distance and two, direction factor, and then obtains candidate matches road set, base
The similitude between GPS track and candidate matches road is calculated in path connected, and then selects similitude the maximum as GPS
The matching road of track；
Step S3 carries out coordinate adjustment to GPS track；
In the step S2, according to distance and two, direction factor establish path adaptation rule for according to distance and two, direction because
Element carries out the building of topological relation and section connected relation, and by the topological relation between the section connected relation and dotted line of building
It is saved in serializing file；
The step S2 further includes following sub-step:
Step S201 calculates error band by the distribution of the GPS positioning probability of error and road width factor, will fall in error band
Gather as the candidate matches section of the GPS point in interior all sections；
Step S202, by the direction difference between the distance between GPS point and section and GPS track and section to candidate
It scores with section, the comprehensive score in candidate matches section is apart from the sum of score and direction score；
Step S203, by the matching result and path connected relationship of a upper GPS point, from the candidate matches section collection to have scored
Optimum Matching road is selected in conjunction；
In the step S201, error band is calculated by formula R=r+w, and then obtain radius as the institute in the border circular areas of R
There is candidate matches road set of the section as the GPS point, wherein w is road width,R is the radius calculated value of candidate matches road set,
C ' and d respectively indicates the axial length of error ellipse,WithFor the standard value of position error,For the initial value of position error；And
The parameters of error ellipse are as follows: WithWherein, θ indicates the angle of error ellipse long axis and y-axis,After indicating unit weight
Error is tested,Indicate covariance；δxAnd δyFor the position error of unit weight, δxyFor the covariance of unit weight.
2. the map-matching method according to claim 1 based on path connected relationship, which is characterized in that the step S1
In, it is removed or interpolation operation, realizes to GPS positioning data respectively for data redundancy, shortage of data and data wander
Pretreatment.
3. the map-matching method according to claim 2 based on path connected relationship, which is characterized in that the step S1
In, when GPS positioning data are there are when polyisomenism, duplicate GPS positioning data are deleted；When vehicle stop and generate two with
When the identical GPS positioning data of upper position coordinates, then when retaining the GPS positioning data at nearest time point and recording the vehicle and stop
Between；When lacking GPS positioning data between two adjacent GPS points, interpolation processing is carried out；It is right when detecting drift data
Drift data carries out delete processing.
4. according to claim 1 to the map-matching method based on path connected relationship described in 3 any one, feature exists
In, in the step S202, if the upright projection point Q ' of GPS point Q to section AB is fallen on the section, the GPS point and section
Distance be the distance between Q and Q ' d (Q, Q ')；It otherwise, is the smaller of GPS point to the section two-end-point distance；Pass through Ws′
It indicates apart from score, then GPS point piWith target road section ejApart from score are as follows:Wherein μ=0, σ etc.
In candidate roads zone radius R；Its direction score passes through formula Wa=cos α is available, wherein note v is vehicle movement side
To g is road bearing mark, and g=0 indicates that two-way traffic road, g=1 are indicated along GPS point number order one-way trip, g
=2 indicate along GPS point number order backward one-way trip, piAnd pi+1Section respectively to be matched is before GPS point number direction
Two-end-point afterwards, then vehicle traffic direction and road direction angle α are as follows:
S ' is distance.
5. the map-matching method according to claim 4 based on path connected relationship, which is characterized in that the step
In S203, pass through formula W=ks·Ws+kα·WaCalculate the comprehensive score in candidate matches section, wherein ksFor distance weighting, kα
For direction weight,
6. the map-matching method according to claim 1 based on path connected relationship, which is characterized in that the step S3
In, the realization process of coordinate adjustment is carried out to GPS track are as follows: when the subpoint of the fitting a straight line endpoint of GPS track falls in candidate
It matches on section, then the distance between projection of the fitting a straight line endpoint is part to be matched；If the projection of fitting a straight line endpoint
Point is fallen in except candidate matches section, then the corresponding end point in candidate matches section should be selected to be matched, coordinate adjustment process warp
Crossing the coordinate obtained after translation, rotation and the geometric transformation of scaling is final matching results.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610107068.9A CN105628033B (en) | 2016-02-26 | 2016-02-26 | A kind of map-matching method based on path connected relationship |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610107068.9A CN105628033B (en) | 2016-02-26 | 2016-02-26 | A kind of map-matching method based on path connected relationship |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105628033A CN105628033A (en) | 2016-06-01 |
| CN105628033B true CN105628033B (en) | 2019-04-02 |
Family
ID=56043183
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610107068.9A Active CN105628033B (en) | 2016-02-26 | 2016-02-26 | A kind of map-matching method based on path connected relationship |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105628033B (en) |
Families Citing this family (41)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105910611A (en) * | 2016-06-27 | 2016-08-31 | 江苏迪纳数字科技股份有限公司 | Road matching method based on matching degree feedback |
| CN106203278B (en) * | 2016-06-28 | 2019-02-26 | 中国人民解放军信息工程大学 | The method and device of two-track road on a kind of extraction map |
| GB201613105D0 (en) * | 2016-07-29 | 2016-09-14 | Tomtom Navigation Bv | Methods and systems for map matching |
| CN107796407A (en) * | 2016-09-06 | 2018-03-13 | 高德软件有限公司 | A kind of map-matching method and device |
| JP6477675B2 (en) * | 2016-12-16 | 2019-03-06 | カシオ計算機株式会社 | Map matching device, map matching method and program |
| CN108267147B (en) * | 2017-01-04 | 2020-08-21 | 北京四维图新科技股份有限公司 | Intelligent generation method and device of navigation data lane communication information, navigation equipment and method |
| CN108286986A (en) * | 2017-01-10 | 2018-07-17 | 厦门雅迅网络股份有限公司 | A kind of vehicle mileage computational methods based on center compensation |
| CN109269514A (en) * | 2017-07-18 | 2019-01-25 | 华为技术有限公司 | The method and apparatus for determining motion profile |
| CN107705545B (en) * | 2017-09-14 | 2021-10-01 | 王淑芳 | Method and system for supplementing track points of key operation vehicles |
| CN107643085B (en) * | 2017-09-18 | 2021-02-12 | 苏州大学 | Path recommendation method and device |
| CN109831736B (en) * | 2017-11-23 | 2022-01-18 | 腾讯科技（深圳）有限公司 | Data processing method and device, server and client |
| CN108562296B (en) * | 2017-12-08 | 2020-09-08 | 北京中交兴路信息科技有限公司 | Road matching method and device based on vehicle satellite positioning data |
| CN109920245A (en) * | 2017-12-12 | 2019-06-21 | 上海宝康电子控制工程有限公司 | Traffic state judging method based on GPS Floating Car information |
| CN108106620B (en) * | 2017-12-20 | 2021-08-24 | 中国科学院深圳先进技术研究院 | Topological road matching method and system and electronic equipment |
| CN108548543B (en) * | 2018-03-27 | 2019-08-02 | 上海识加电子科技有限公司 | Vehicle route management method and device |
| CN108763558B (en) * | 2018-05-25 | 2020-12-18 | 武汉大学 | Crowdsourcing map road quality improvement method based on map matching |
| CN108959414B (en) * | 2018-06-07 | 2021-06-08 | 同济大学 | Method for extracting missing road from navigation data points based on machine learning |
| CN108763817B (en) * | 2018-06-11 | 2022-06-21 | 厦门理工学院 | Electric underground pipe network matching method based on least square modeling |
| CN109029472A (en) * | 2018-07-10 | 2018-12-18 | 天津大学 | Map-matching method based on low sampling rate GPS track point |
| CN109141405B (en) * | 2018-08-06 | 2021-12-21 | 中国科学院光电研究院 | Vehicle geomagnetic matching positioning method and system in road network environment |
| CN109147323A (en) * | 2018-08-28 | 2019-01-04 | 华南理工大学 | A kind of vehicle GPS data processing method for highway passenger and freight transportation indicator-specific statistics |
| CN109523781B (en) * | 2018-10-10 | 2020-07-28 | 中国电子科技集团公司第二十八研究所 | Intersection prediction method based on satellite positioning |
| CN110176139B (en) * | 2019-02-21 | 2021-01-05 | 淮阴工学院 | DBSCAN + -based road congestion identification visualization method |
| CN111721306B (en) * | 2019-03-20 | 2022-07-05 | 北京嘀嘀无限科技发展有限公司 | Road matching method and device, electronic equipment and readable storage medium |
| CN109827581B (en) * | 2019-03-28 | 2020-03-13 | 北京三快在线科技有限公司 | Map matching method and device |
| CN110046683A (en) * | 2019-04-24 | 2019-07-23 | 安徽建筑大学 | A kind of production of prefabricated components, dispatching and assembly integrated information management system |
| CN110686686B (en) * | 2019-06-04 | 2020-10-02 | 滴图(北京)科技有限公司 | System and method for map matching |
| CN110427360B (en) * | 2019-06-28 | 2022-02-11 | 东软集团股份有限公司 | Processing method, processing device, processing system and computer program product of trajectory data |
| CN110567463A (en) * | 2019-08-16 | 2019-12-13 | 中国人民解放军63921部队 | matching method and device of electronic map |
| CN112444834A (en) | 2019-08-29 | 2021-03-05 | 华为技术有限公司 | Positioning method and electronic equipment |
| CN110856119B (en) * | 2019-11-08 | 2021-07-13 | 中国测绘科学研究院 | Port vehicle track deviation rectifying method based on geometric vector |
| CN111307162B (en) * | 2019-11-25 | 2020-09-25 | 奥特酷智能科技(南京)有限公司 | Multi-sensor fusion positioning method for automatic driving scene |
| CN113132907B (en) * | 2019-12-27 | 2022-11-08 | 中国移动通信集团浙江有限公司 | Pedestrian flow distribution identification method and device, computing device and storage device |
| CN111127950A (en) * | 2019-12-27 | 2020-05-08 | 北京万集智能网联技术有限公司 | Vehicle collision early warning processing method and device |
| CN111309780B (en) * | 2020-01-21 | 2023-04-25 | 腾讯云计算（北京）有限责任公司 | Track data processing method and device |
| CN112559660B (en) * | 2020-12-11 | 2022-06-17 | 腾讯科技（深圳）有限公司 | Road data processing method and device, electronic equipment and storage medium |
| CN113514860A (en) * | 2021-04-23 | 2021-10-19 | 交通运输部公路科学研究所 | Road matching method and device |
| CN112987052B (en) * | 2021-04-27 | 2021-08-10 | 中南大学 | Rapid graph matching method based on road network section classification |
| CN113639757B (en) * | 2021-07-29 | 2023-08-25 | 上海交通大学 | Map matching method and system based on bidirectional scoring model and backtracking correction mechanism |
| CN113959452A (en) * | 2021-10-22 | 2022-01-21 | 上海经达信息科技股份有限公司 | Map matching method, system and terminal based on urban road network |
| CN114234991A (en) * | 2021-11-25 | 2022-03-25 | 中国南方电网有限责任公司 | Navigation path planning method and device, computer equipment and storage medium |
Family Cites Families (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP4377284B2 (en) * | 2004-06-02 | 2009-12-02 | 株式会社ザナヴィ・インフォマティクス | Car navigation system |
| WO2008117787A1 (en) * | 2007-03-27 | 2008-10-02 | Nec Corporation | Map matching system, map matching method and program |
| CN101201255A (en) * | 2007-12-21 | 2008-06-18 | 四川川大智胜软件股份有限公司 | Vehicle combined navigation system based on intelligent navigation algorithm |
| CN101545788B (en) * | 2009-05-14 | 2011-07-27 | 四川长虹电器股份有限公司 | Vehicle positioning implementing method |
| CN102147261B (en) * | 2010-12-22 | 2013-07-03 | 广州通易科技有限公司 | Method and system for map matching of transportation vehicle GPS (Global Position System) data |
| CN102147258B (en) * | 2010-12-24 | 2012-12-26 | 清华大学 | Feedback-mechanism-based vehicle navigation method and system |
| CN102183256B (en) * | 2011-02-28 | 2012-12-26 | 重庆大学 | Map matching method for marching fleet |
| CN102435197B (en) * | 2011-08-05 | 2014-10-15 | 刘建勋 | MBR (Master Boot Record)-based GPS (Global Position System) track map matching method |
| CN102521973B (en) * | 2011-12-28 | 2016-03-30 | 昆明理工大学 | A kind of mobile phone switches the road matching method of location |
| CN103123704B (en) * | 2013-01-21 | 2016-02-24 | 浙江工业大学 | A kind of Logistics Distribution Method based on rich network attribute road network |
| CN103363999B (en) * | 2013-07-30 | 2016-09-14 | 武汉大学 | A kind of alternative manner generating road network stroke |
| CN105021202B (en) * | 2014-04-29 | 2018-09-28 | 高德软件有限公司 | A kind of map-matching method and equipment |
| CN104240260B (en) * | 2014-10-09 | 2017-01-18 | 武汉大学 | Junction identification based intelligent road extraction method |
| CN104330089B (en) * | 2014-11-17 | 2017-12-29 | 东北大学 | A kind of method that map match is carried out using history gps data |
| CN104515528B (en) * | 2014-12-22 | 2019-01-29 | 厦门雅迅网络股份有限公司 | Single-point map-matching method based on section cumulative probability |
| CN105303839B (en) * | 2015-12-04 | 2017-12-08 | 中国科学院遥感与数字地球研究所 | The Forecasting Methodology and device in potential jam road crosspoint |
-
2016
- 2016-02-26 CN CN201610107068.9A patent/CN105628033B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN105628033A (en) | 2016-06-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105628033B (en) | A kind of map-matching method based on path connected relationship | |
| Hsueh et al. | Map matching for low-sampling-rate GPS trajectories by exploring real-time moving directions | |
| Yuan et al. | An interactive-voting based map matching algorithm | |
| EP1946044B1 (en) | System and method for identifying road features | |
| CN104197945B (en) | Global voting map matching method based on low-sampling-rate floating vehicle data | |
| CN110427360A (en) | Processing method, processing unit, processing system and the computer program product of track data | |
| CN102183256B (en) | Map matching method for marching fleet | |
| CN110095127A (en) | A kind of hidden Markov model map-matching method based on segmentation | |
| CN106446960A (en) | Dynamic hot degree road network constructing method with density-based clustering time quantum dividing function | |
| CN110285817B (en) | Complex road network map matching method based on self-adaptive D-S evidence theory | |
| Li et al. | Robust inferences of travel paths from GPS trajectories | |
| Blazquez et al. | Simple map-matching algorithm applied to intelligent winter maintenance vehicle data | |
| CN105928529A (en) | Map-matching algorithm for combining multiple evidences | |
| Schulze et al. | Map-matching cell phone trajectories of low spatial and temporal accuracy | |
| Kong et al. | A scenario-based map-matching algorithm for complex urban road network | |
| CN109121094A (en) | The pretreatment of pseudo-code signaling data and Trip chain recognition methods | |
| Vuurstaek et al. | GTFS bus stop mapping to the OSM network | |
| CN106289281A (en) | A kind of double mode map-matching method theoretical based on three evidence DS | |
| CN113295173A (en) | Map matching method for annular road section | |
| de Sousa et al. | A map matching based framework to reconstruct vehicular trajectories from gps datasets | |
| CN111123333B (en) | Vehicle track positioning method fusing bayonet and GPS data | |
| CN116597666A (en) | Urban road network non-detector road section flow real-time estimation method based on transfer learning | |
| Yang et al. | The research on real-time map-matching algorithm | |
| Sun et al. | Excessive-emission vehicles real-time track matching algorithm based on road network topology and weights | |
| de Sousa et al. | A novel scalable framework to reconstruct vehicular trajectories from unreliable gps datasets |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |