CN107328423A - Bend recognition methods and its system based on map datum - Google Patents
Bend recognition methods and its system based on map datum Download PDFInfo
- Publication number
- CN107328423A CN107328423A CN201610277789.4A CN201610277789A CN107328423A CN 107328423 A CN107328423 A CN 107328423A CN 201610277789 A CN201610277789 A CN 201610277789A CN 107328423 A CN107328423 A CN 107328423A
- Authority
- CN
- China
- Prior art keywords
- bend
- node
- road
- line segment
- deflection difference
- 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
- 238000000034 method Methods 0.000 title claims abstract description 27
- 239000012141 concentrate Substances 0.000 claims description 3
- 230000000694 effects Effects 0.000 abstract description 6
- 238000010586 diagram Methods 0.000 description 7
- 230000001133 acceleration Effects 0.000 description 5
- 238000005516 engineering process Methods 0.000 description 5
- 239000000203 mixture Substances 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 2
- 230000003467 diminishing effect Effects 0.000 description 2
- 238000000605 extraction Methods 0.000 description 2
- 230000000007 visual effect Effects 0.000 description 2
- 101100366707 Arabidopsis thaliana SSL11 gene Proteins 0.000 description 1
- 101100366710 Arabidopsis thaliana SSL12 gene Proteins 0.000 description 1
- 101100366711 Arabidopsis thaliana SSL13 gene Proteins 0.000 description 1
- 206010057315 Daydreaming Diseases 0.000 description 1
- 206010057855 Hypotelorism of orbit Diseases 0.000 description 1
- 101100366561 Panax ginseng SS11 gene Proteins 0.000 description 1
- 101100366562 Panax ginseng SS12 gene Proteins 0.000 description 1
- 101100366563 Panax ginseng SS13 gene Proteins 0.000 description 1
- 206010039203 Road traffic accident Diseases 0.000 description 1
- 101000662518 Solanum tuberosum Sucrose synthase Proteins 0.000 description 1
- 230000006978 adaptation Effects 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000001427 coherent effect Effects 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 230000000977 initiatory effect Effects 0.000 description 1
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)
- Traffic Control Systems (AREA)
Abstract
The invention discloses a kind of bend recognition methods based on map datum and its system, method includes：Vehicle location is matched in map, road where vehicle is obtained；The road circuit node in the road circuit node in default first distance in vehicle location rear on road and the second distance of vehicle location front preset is obtained, node linked list is obtained；Using each road circuit node as starting point in calculate node chained list, one of circuit node is the deflection of the line segment of terminal below；Calculate the deflection difference of line segment and its previous line segment；Deflection difference is designated as to the deflection difference of the road circuit node of correspondence line segment starting point；According to deflection difference, bend node is obtained, bend chained list is obtained.The present invention is combined judgement with map datum, when the night of rainy day, insufficient light even in low visibility or pavement of road do not have any feature, also can Real time identification go out the degree of crook of vehicle front bend so that bend identification can not carried out with round-the-clock by road surface effect.
Description
Technical field
The present invention relates to traffic safety field, more particularly to a kind of bend recognition methods based on map datum and its it is
System.
Background technology
Vehicle driver due to absent minded or be unfamiliar with the reasons such as road ahead, when front occur bend without
Often lead to great traffic accident when taking the necessary measures.Therefore, the bend that fast and effeciently identification road ahead is present
And estimate the degree of crook gone off the curve one of important technology as current automobile intelligent safety DAS (Driver Assistant System).
Traditional bend identification technology mainly installs ccd image sensor in vehicle front and obtains visual information, obtains
A series of digital picture, recognizes the curve that road lines are constituted, thus by methods such as mathematical modeling and image procossings
Detect and go off the curve feature and follow-up early warning processing is carried out with this.However, the technology receives the imaging sensor visual field and Road
The limitation in face.When the rainy day of low visibility or the night of insufficient light or pavement of road do not have any feature, using this skill
Art will be difficult to effectively identify road lines, so as to cause recognition failures.
In Publication No. CN1796940 Chinese patent open file, it is proposed that a kind of curved road safety of navigation system
Method for early warning, it learns that the current coordinate points of vehicle, and inertia measuring equipment learn vehicle traveling direction using GPS, and utilizes
The current coordinate points of vehicle take forward a segment distance as estimating distance and taking out reduced coordinates point, according to current speed and electronics
The safe speed value for the reduced coordinates point that map is prestored is compared, and is warned if speed exceedes safe speed value.Should
Scheme is to estimate in electronic map the curvature of all road circuit nodes in advance and calculate safe speed value, then write-in ground picture and text
Part, because to calculate all road circuit nodes, data volume is huge, can reduce efficiency, and for the calculating of bend curvature,
Only extract three nodes therein to be calculated, the degree of accuracy is relatively low.
The content of the invention
The technical problems to be solved by the invention are：A kind of bend recognition methods based on map datum is provided and its is
System, is identified and improves the accuracy of bend curvature estimation to the bend of vehicle front in real time.
In order to solve the above-mentioned technical problem, the technical solution adopted by the present invention is：A kind of bend based on map datum is known
Other method, including：
Vehicle location is matched in map, road where vehicle is obtained；
Obtain pre- in front of the road circuit node and vehicle location in default first distance in vehicle location rear on the road
If second distance in road circuit node, obtain node linked list；
Calculate using each road circuit node as starting point in the node linked list, one of circuit node is the side of the line segment of terminal below
To angle；
Calculate the deflection difference of the line segment and its previous line segment；
The deflection difference is designated as to the deflection difference of the road circuit node of the correspondence line segment starting point；
According to the deflection difference, bend node is obtained, bend chained list is obtained.
The invention further relates to a kind of bend identifying system based on map datum, including：
Matching module, for matching vehicle location in map, obtains road where vehicle；
First obtains module, for obtaining the road circuit node in default first distance in vehicle location rear on the road
Road circuit node with the second distance of vehicle location front preset, obtains node linked list；
First computing module, for calculating using each road circuit node as starting point in the node linked list, a road section below
Point is the deflection of the line segment of terminal；
Second computing module, the deflection difference for calculating the line segment and its previous line segment；
Mark module, the direction angular difference of the road circuit node for the deflection difference to be designated as to the correspondence line segment starting point
Value；
Second obtains module, for according to the deflection difference, obtaining bend node, obtaining bend chained list.
The beneficial effects of the present invention are：By being extracted in map datum in the certain limit of current vehicle location front and back
Road node linked list, according to road circuit node move towards deflection deviation distinguish go off the curve or straight way, extract bend node,
Obtain bend chained list；The present invention is combined judgement with map datum, the night of rainy day, insufficient light even in low visibility or road
When road road surface does not have any feature, also can Real time identification go out the degree of crook of vehicle front bend so that bend identification can be complete
Do not carried out by road surface effect weather；Meanwhile, only obtain a range of road node data in vehicle location front and back and come
Bend identification is carried out, efficiency is also ensure that while real-time bend is recognized.
Brief description of the drawings
Fig. 1 is a kind of flow chart of the bend recognition methods based on map datum of the present invention；
Fig. 2 is the method flow diagram of the embodiment of the present invention one；
Fig. 3 is the method flow diagram of the embodiment of the present invention two；
Fig. 4 is the method flow diagram of the embodiment of the present invention three；
Fig. 5 is the method flow diagram of the embodiment of the present invention four；
The road circuit node schematic diagram of Fig. 6 roads where the vehicle of the embodiment of the present invention five；
Fig. 7 is a kind of structural representation of the bend identifying system based on map datum of the present invention；
Fig. 8 is the system structure diagram of the embodiment of the present invention six.
Label declaration：
1st, matching module；2nd, first module is obtained；3rd, the first computing module；4th, the second computing module；5th, mark module；6、
Second obtains module；7th, the 3rd computing module；8th, removing module；9th, the first update module；10th, the second update module；11st, obtain
Module；12nd, the 4th computing module；
101st, first acquisition unit；102nd, second acquisition unit；103rd, the first judging unit；104th, culling unit；105、
First computing unit；106th, unit is chosen；
601st, the first division unit；602nd, first unit is added；603rd, first unit is deleted；604th, first unit is obtained；
605th, second unit is added；606th, Traversal Unit；607th, the second computing unit；608th, the second division unit；609th, the 3rd calculate
Unit；610th, second unit is deleted；
1201st, projecting cell；1202nd, rotary unit；1203rd, translation unit；1204th, the 4th computing unit；1205th, it is fitted
Unit；1206th, the 5th computing unit.
Embodiment
To describe technology contents, the objects and the effects of the present invention in detail, below in conjunction with embodiment and coordinate attached
Figure is explained in detail.
The design of most critical of the present invention is：Identified according to the deflection deviation that road circuit node is moved towards before vehicle location
The bend of side.
Referring to Fig. 1, a kind of bend recognition methods based on map datum, including：
Vehicle location is matched in map, road where vehicle is obtained；
Obtain pre- in front of the road circuit node and vehicle location in default first distance in vehicle location rear on the road
If second distance in road circuit node, obtain node linked list；
Calculate using each road circuit node as starting point in the node linked list, one of circuit node is the side of the line segment of terminal below
To angle；
Calculate the deflection difference of the line segment and its previous line segment；
The deflection difference is designated as to the deflection difference of the road circuit node of the correspondence line segment starting point；
According to the deflection difference, bend node is obtained, bend chained list is obtained.
It was found from foregoing description, the beneficial effects of the present invention are：By GPS and map datum, even in low visibility
Rainy day, the night of insufficient light or pavement of road when there is no any feature, also can Real time identification go out the curved of vehicle front bend
Qu Chengdu so that bend identification can not carried out with round-the-clock by road surface effect.
Further, described " vehicle location being matched in map, obtain road where vehicle " is specially：
Obtain the current gps data of vehicle；
According to the coordinate information in the gps data, the candidate roads in vehicle periphery preset range are obtained, first is obtained
Candidate roads collection；
Judge the road and the connectedness of the candidate roads of vehicle last time matching；
Concentrated in first candidate roads and reject disconnected candidate roads, obtain the second candidate roads collection；
Calculate the phase that second candidate roads concentrate each candidate roads and the vector position information in the gps data
Like degree；
The minimum candidate roads of the similarity are chosen as road where vehicle.
Seen from the above description, according to gps data, choosing similar with the candidate roads of the upper path connected once matched
The minimum road of degree, can correctly match the road where vehicle, improve the path adaptation degree of accuracy.
Further, after described " obtaining node linked list ", further comprise：
Calculate the distance between two nodes adjacent in the node linked list；
If the distance is less than default 3rd distance, any node in described two nodes is deleted.
Seen from the above description, it can prevent from causing at calculated direction angle due to two nodes got too close in map datum
When produce larger error, improve the accuracy rate of bend identification.
Further, after described " the deflection difference of the line segment and its previous line segment ", further comprise：
If the deflection difference S ' of line segment described in one and its previous line segment is more than 180 degree, the deflection difference is updated
For | S ' | -360；
If the deflection difference S ' of line segment described in one and its previous line segment is less than -180 degree, the deflection difference is updated
For 360- | S ' |.
Seen from the above description, because in map datum, adjacent two lines section deflection change will not very greatly, therefore
Think that current line segment deflection is located at 0 ° of axle two with upper line segment direction angle when the absolute value of deflection difference is more than 180 °
Side, by updating deflection difference, prevents from causing error in judgement, it is ensured that the accuracy of bend identification.
Further, described " according to the deflection difference, obtain bend node, obtain bend chained list " is specially：
It is, on the occasion of, negative value or zero situation, sequentially to divide in the node linked list except first road section according to deflection difference
Road circuit node outside point, obtains the node packet of sequential；
First road circuit node that the node is grouped is added in previous node packet；
Delete the node packet that the deflection difference of remaining road circuit node in addition to extreme trace circuit node is zero；
Obtain being grouped the bend chained list constituted by remaining node, node packet one bend of correspondence described in one.
Seen from the above description, direct north is set as 0 ° of axle, is clockwise positive direction；When the side of one of circuit node
To angular difference value be negative value when, illustrate that the line segment direction angle compared with of the line segment using the road circuit node as starting point is diminishing, road is
Turning left；When one circuit node deflection difference be on the occasion of when, illustrate the line compared with of the line segment using the road circuit node as starting point
Section deflection is becoming big, and road is to turn right；When the deflection difference of the road circuit node is 0, illustrate that road does not have direction change
Change.
Further, it is described " to obtain being grouped the bend chained list constituted by remaining node, node packet correspondence one described in one
After bend ", further comprise：
If the node packet is only comprising two road circuit nodes, by the previous road adjacent with described two road circuit nodes
Node is added in the node packet.
Seen from the above description, because one section of camber line is at least made up of three points, therefore, if node packet is a bend
Contains only two road circuit nodes, then need the adjoint point before 2 points being added in the bend using ensure bend number of nodes as
More than three.
Further, it is described " to obtain being grouped the bend chained list constituted by remaining node, node packet correspondence one described in one
After bend ", further comprise：
Travel through the bend node of each bend in the bend chained list；
If the corresponding deflection difference of a bend node is less than default first difference, the bend node is calculated with
The distance of one bend node；
If the distance is more than default 4th distance, according to the bend node, the bend node is corresponding
Bend is divided into two bends；
The discrepancy angle absolute difference of each bend is calculated respectively；
If the discrepancy angle absolute difference of a bend is less than default second difference, by the bend from bend chain
Deleted in table.
Seen from the above description, by being modified to the bend in bend chained list, reducing the error of map datum causes
Error in judgement, further improve bend identification accuracy rate.
Further, after described " according to the deflection difference, obtain bend node, obtain bend chained list ", one is entered
Step includes：
If vehicle is in a corresponding bend in the bend chained list after preset time, the bend is obtained corresponding curved
Road set of node；
According to the bend set of node, carry out curve fitting, calculate the radius of curvature of the bend.
Further, described " according to the bend set of node, carrying out curve fitting, calculate the radius of curvature of the bend "
Specially：
The bend node of the bend is projected in rectangular coordinate system；
Line direction by the bend node according to bend first and last node is rotated to the X-axis of the parallel rectangular coordinate system
Direction；
The bend node is moved to the origin of the rectangular coordinate system according to bend first node；
Calculate relative coordinate of each bend node with respect to bend first node；
According to the relative coordinate, quadratic polynomial least square fitting is carried out to the bend node, curve is obtained
Fitting coefficient；
According to the fitting coefficient, the radius of curvature of the bend is calculated.
Seen from the above description, the bend set of node in bend chained list carries out curve fitting, and calculates the curvature of bend
Radius, safe early warning is carried out available for bend speed, improves travel safety.
Fig. 7 is refer to, the invention further relates to a kind of bend identifying system based on map datum, including：
Matching module, for matching vehicle location in map, obtains road where vehicle；
First obtains module, for obtaining the road circuit node in default first distance in vehicle location rear on the road
Road circuit node with the second distance of vehicle location front preset, obtains node linked list；
First computing module, for calculating using each road circuit node as starting point in the node linked list, a road section below
Point is the deflection of the line segment of terminal；
Second computing module, the deflection difference for calculating the line segment and its previous line segment；
Mark module, the direction angular difference of the road circuit node for the deflection difference to be designated as to the correspondence line segment starting point
Value；
Second obtains module, for according to the deflection difference, obtaining bend node, obtaining bend chained list.
Embodiment one
Fig. 2 is refer to, embodiments of the invention one are：A kind of bend recognition methods based on map datum, can be used for into
Row bend Curvature Estimate and bend speed safe early warning, specifically include following steps：
S1：Vehicle location is matched in map, road where vehicle is obtained；Vehicle is travelled on road, therefore, car
Position can be matched to bar none on a certain bar road in map.
S2：Obtain in front of the road circuit node and vehicle location in default first distance in vehicle location rear on the road
Road circuit node in default second distance, obtains node linked list；Road is constituted in the form of node on map, and road
Road in road network has certain topological relation；Typically, vehicle typically has coherent in higher level road driving
Property, i.e. current location are identical with front and back category of roads, road name is identical, and current location front and back one is searched out with this
In set a distance, such as 2000m, backward 500m road circuit node, and constitute road node linked list forward.
S3：Calculate using each road circuit node as starting point in the node linked list, one of circuit node is the line segment of terminal below
Deflection；Direct north is set as 0 ° of axle, is clockwise positive direction.
S4：Calculate the deflection difference of the line segment and its previous line segment；Due in map datum, adjacent two lines
Section deflection change will not very greatly, therefore when deflection difference absolute value be more than 180 ° when think current line segment deflection and
Upper line segment direction angle is located at 0 ° of axle both sides, therefore, if the deflection difference S ' of line segment described in one and its previous line segment is more than 180
Degree, then updating the deflection difference is | S ' | -360；If the deflection difference S ' of line segment described in one and its previous line segment is less than-
180 degree, then update the deflection difference for 360- | S ' |；By updating deflection difference, prevent from causing error in judgement, it is ensured that
The accuracy of bend identification.
S5：The deflection difference is designated as to the deflection difference of the road circuit node of the correspondence line segment starting point；In node
In chained list, except first and last node, remaining all road circuit node corresponds to a deflection difference, and the deflection difference is with described
Road circuit node is the line segment direction angle of starting point and using the road circuit node as the difference at the line segment direction angle of terminal.Road in node linked list
Circuit node is, according to road direction of advance number consecutively, when the deflection difference of one of circuit node is negative value, to illustrate with this
Road circuit node is diminishing at a line segment direction angle for the line segment of starting point compared with, and road is to turn left；When one of circuit node deflection
Difference be on the occasion of when, illustrate the line segment using the road circuit node as starting point compared with a line segment direction angle become greatly, road is on the right side
Turn；When the deflection difference of the road circuit node is 0, illustrate that road does not have direction change.
S6：It is, on the occasion of, negative value or zero situation, sequentially to divide in the node linked list except first road according to deflection difference
Road circuit node outside node, obtains the node packet of sequential.Such as the corresponding direction of road circuit node in a node linked list
Angular difference value is respectively 0,8,15,6,0,0, -5, -10, -7,4,9,0, then be sequentially divided into (8,15,6), (0,0), (- 5, -10, -
7), (4,9), (0) five node packet.
S7：First road circuit node that the node is grouped is added in previous node packet；The in node packet
One road circuit node is both the starting point that the node is grouped corresponding section, is also the terminal in a upper section.
S8：Whether the deflection difference of remaining road circuit node is zero in addition to extreme trace circuit node in decision node packet, if
It is then to perform step S9.
S9：The node packet is deleted, obtains being grouped the bend chained list constituted by remaining node, node described in one is grouped
One bend of correspondence.
S10：Whether the remaining node packet is judged only comprising two road circuit nodes, if so, step S11 is then performed,
If it is not, then performing step S12.Because one section of camber line is at least made up of three points, therefore, if a bend contains only two roads
Node is, it is necessary to the adjoint point before 2 points is added in the bend to ensure the road number of nodes of bend as more than three.
S11：The previous road circuit node adjacent with described two road circuit nodes is added in the node packet；Perform step
Rapid S12.
S12：Correct the bend in the bend chained list；Due to the error of map datum, there can be following two kinds of situations：
Two sections of bends are extracted into one section of bend by a, mistake；Although extraction node packet in addition to extreme trace circuit node direction
Angular difference value is all just or is all negative, but centre there may be the road circuit node that deflection difference gets too close to 0, the reflection of these nodes
It is the straight way that occurs in bend on to map, due to the error of map datum, fails the deflection difference of road circuit node at this
0 is calculated as, fails to disconnect in this place.
Straight way is extracted into bend by b, mistake；Although direction angular difference value is same in addition to extreme trace circuit node in the node packet of extraction
For just or be all negative, but the deflection difference of all road circuit nodes may all get too close to 0, in this bend reflection to map i.e.
For straight way, due to the error of map datum, fail to be calculated as 0 in calculated direction angular difference value.
Therefore need to be modified the bend in bend chained list, judgement is lost caused by can reducing the error of map datum
By mistake, the accuracy rate of bend identification is improved.
Preferably, between step S2 and step S3, can also first it calculate between two nodes adjacent in the node linked list
Distance；If the distance is less than default 3rd distance, such as 30m deletes any node in described two nodes.Because ground
Error is there is in diagram data, the point of hypotelorism means that at calculated direction angle larger error can be produced, by deleting
Near road circuit node is crossed, error is reduced, the accuracy rate of bend identification is improved.
The present embodiment in map datum by extracting a range of road circuit node chain in current vehicle location front and back
Table, the deflection deviation moved towards according to road circuit node is distinguished and gone off the curve or straight way, rainy day, light even in low visibility
When not enough night or pavement of road does not have any feature, also can Real time identification go out the degree of crook of vehicle front bend so that
Bend identification can not carried out with round-the-clock by road surface effect；Meanwhile, only obtain vehicle location front and back a range of
Road node data carries out bend identification, while real-time bend is recognized also ensure that efficiency.
Embodiment two
Fig. 3 is refer to, the present embodiment is the further expansion of embodiment one, according to the bend chained list obtained in embodiment one
Further analysis, carries out bend curvature estimation, for bend speed safe early warning；Something in common is not repeated, and difference is, also
Comprise the following steps：
S13：According to vehicle present speed and acceleration, the distance of vehicle traveling is calculated in preset time and in preset time
Instantaneous velocity afterwards；The motion model of vehicle can be reduced to even plus (subtracting) fast motion model, according to vehicle present speed v0With
The acceleration a that the acceleration transducer of vehicle is obtained, calculate that t second rear vehicles travel apart from s=v0t+1/2at2, after t seconds
Instantaneous velocity v=v0+ at, wherein, t is the default time；When acceleration is 0, vehicle is to move with uniform velocity.
S14：According to the distance, the location of preset time rear vehicle is determined in node linked list.
S15：Judge whether vehicle is in a corresponding bend in the bend chained list after preset time, if so, then holding
Row step S16.
S16：Obtain the corresponding bend set of node of the bend.
S17：The bend node of the bend is projected in rectangular coordinate system.
S18：Line direction by the bend node according to bend first and last node is rotated to the parallel rectangular coordinate system
X-direction.
S19：The bend node is moved to the origin of the rectangular coordinate system according to bend first node.
S20：Calculate relative coordinate of each bend node with respect to bend first node.
S21：According to the relative coordinate, quadratic polynomial least square fitting is carried out to the bend node, song is obtained
The fitting coefficient of line；For the model of curve, second-degree parabola model is used, for quadratic polynomial Y (X)=b0+b1X+
b2X2, it is known that n discrete point (Xi,Yi), i=1,2 ... n can be in the hope of:
Wherein：
S22：According to the fitting coefficient, the radius of curvature of the bend is calculated；The radius of curvature value of curve is on curve
Every bit it is all different；For judging bend, can most represent its degree of crook for radius of curvature minimum at point, i.e.,
Radius of curvature at the most curved point of bend, the point puts horizontal axis etc. minute 100 by the way of estimation in curve first point and end
It is individual, the radius of curvature of this 100 points is calculated, minimum value therein is sought.The calculation formula of radius of curvature is：
S23：According to the radius of curvature, calculating vehicle can be with the maximal rate of safety bend；For highway,
Its appraising model is：Car V=14.741R0.3194, truck V=4.941R0.4345；For Class II highway, its appraising model
For：Car V=14.391R0.2757, truck V=8.503R0.3168。
S24：According to the instantaneous velocity after the preset time and the max speed, safe early warning is carried out；If step S13
Instantaneous velocity after the t seconds calculated is more than the maximal rate that step S23 is calculated, then carries out early warning prompting to driver.
Radius of curvature estimation is carried out to the bend identified, and for vehicle bend safe early warning, traffic safety can be improved
Property.
Embodiment three
Fig. 4 is refer to, the present embodiment is the step S1 of embodiment one specific implementation, is comprised the following steps：
SS11：Obtain the current gps data of vehicle；The gps data includes coordinate information and vector position information.
SS12：According to the coordinate information in the gps data, the candidate roads in vehicle periphery preset range are obtained, are obtained
To the first candidate roads collection.
SS13：Judge the road and the connectedness of the candidate roads of vehicle last time matching；Matched with the last time
Road distal point is initiating searches point, preset search number of times, such as 6 times, pointed by the searching times inner iteration initial search point
Child node, the child node of child node, the like, if the node searched in default searching times is composition candidate road
The node of road in the set of road, then the candidate roads are with originating path connected；It is on the contrary then do not connect.
SS14：Concentrated in first candidate roads and reject disconnected candidate roads, obtain the second candidate roads collection.
SS15：Calculating second candidate roads concentrates each candidate roads to believe with the vector position in the gps data
The similarity of breath；Vector position information is directive positional information, and similarity can be according to formula S=waa+wdD is calculated and obtained,
Wherein, a, d are respectively the differential seat angle and distance of current vector positional information and candidate roads, waAnd wdFor the differential seat angle and away from
From weighted value.
SS16：The minimum candidate roads of the similarity are chosen as road where vehicle.
Example IV
Fig. 5 is refer to, the present embodiment is the step S12 of embodiment one specific implementation, is comprised the following steps：
SS121：Travel through the bend node of each bend in the bend chained list.
SS122：Judge whether the corresponding deflection difference of bend node is less than default first difference, if so, then performing
Step SS123；Preferably, first difference is 5 °.
SS123：Calculate the distance of the bend node and next bend node.
SS124：Judge whether the distance is more than default 4th distance, if so, then performing step SS125；Preferably,
4th distance is 250m.
SS125：According to the bend node, the corresponding bend of the bend node is divided into two bends；Will institute
State the corresponding bend of bend node to disconnect in the bend node, be divided into two bends.
SS126：The discrepancy angle absolute difference of each bend is calculated respectively；The angular difference value that comes in and goes out is constitute bend first
Line segment (the first bend node using bend as starting point, using second bend node of bend as the line segment of terminal) and composition bend
The last item line segment (the penultimate bend node using bend as starting point, using the last bend node of bend as the line segment of terminal)
The difference of deflection.
SS127：Judge whether the discrepancy angle absolute difference of bend is less than default second difference, if so, then holding
Row step SS128；Preferably, second difference is 5 °.
SS128：The bend is deleted from bend chained list.
The present embodiment to the bend in bend chained list by being modified, and judgement is lost caused by reducing the error of map datum
By mistake, the accuracy rate of bend identification is further improved.
Embodiment five
The present embodiment is an embodiment of above-described embodiment.
As shown in fig. 6, matching vehicle location in map, the road in figure has been matched, vehicle location front and back are obtained
Gong19Ge roads circuit node, numbering is respectively 0-18, obtains the node linked list of 19 roads circuit node composition；Calculate road two-by-two
The distance between node, the distance of setting the 3rd is 30m, calculates the distance between No. 10 and No. 11 Liang Ge roads circuit nodes deficiency
30m, then delete No. 11 road circuit nodes in node linked list；The corresponding deflection difference of remaining road circuit node is calculated, such as the institute of table 1
Show.
| Road node serial number | Coordinate | Deflection difference |
| 0 | 118.117732,24.433752 | 0 |
| 1 | 118.119032,24.433921 | 3 |
| 2 | 118.120414,24.434047 | 0 |
| 3 | 118.124017,24.434338 | 7 |
| 4 | 118.125200,24.434321 | 17 |
| 5 | 118.126510,24.433952 | 7 |
| 6 | 118.129280,24.432789 | -8 |
| 7 | 118.130106,24.432559 | -9 |
| 8 | 118.131882,24.432346 | -10 |
| 9 | 118.132719,24.432372 | -10 |
| 10 | 118.134160,24.432637 | -19 |
| 12 | 118.135618,24.433418 | -19 |
| 13 | 118.136415,24.434256 | -12 |
| 14 | 118.136968,24.435178 | -1 |
| 15 | 118.143383,24.446467 | 12 |
| 16 | 118.144329,24.447511 | 10 |
| 17 | 118.144898,24.447958 | 8 |
| 18 | 118.145577,24.448349 | 0 |
Table 1
It is, on the occasion of, negative value or zero situation, sequentially to divide and first and last road is removed in the node linked list according to deflection difference
Road circuit node outside node, gets the node packet of sequential, as shown in table 2.
| Node packet numbering | Road node serial number |
| 0 | 1 |
| 1 | 2 |
| 2 | 3、4、5 |
| 3 | 6、7、8、9、10、12、13、14 |
| 4 | 15、16、17 |
| 5 | 18 |
Table 2
First road circuit node that the node is grouped is added in previous node packet, as shown in table 3.
| Node packet numbering | Road node serial number |
| 0 | 1、2 |
| 1 | 2、3 |
| 2 | 3、4、5、6 |
| 3 | 6、7、8、9、10、12、13、14、15 |
| 4 | 15、16、17、18 |
| 5 | 18 |
Table 3
The node packet that the deflection difference of remaining road circuit node in addition to extreme trace circuit node is zero is deleted, that is, deletes No. 1
Node with No. 5 is grouped；If there is the node packet for only including two road circuit nodes in the packet of remaining node, will with it is described
The adjacent previous road circuit node of two road circuit nodes is added in the node packet, i.e., the road circuit node of No. 0 is added into No. 0
Node is grouped, and obtains bend chained list, node packet one bend of correspondence, and it is absolute to calculate the discrepancy angular difference value of each bend
Value, as shown in table 4.
Table 4
The bend in bend chained list is corrected, No. 0 bend that the angle absolute difference that comes in and goes out is less than 5 ° is deleted；With reference to table 1, due to
The absolute value of No. 14 road node direction angular difference values is less than 5 °, it is therefore desirable to calculates the distance of itself and No. 15 road circuit nodes, calculates
Distance value is 1415m, is disconnected more than set threshold value the 4th apart from 250m, therefore No. 2 bends from No. 14 points.
Assuming that preset time is 10 seconds, it is No. 2 bends to calculate position of the vehicle after 10 seconds, therefore is fitted the bend
Conic section, the Curve Resolution formula after fitting is：Y (X)=198-0.8531X+0.0009214X2, the radius of curvature of the bend
It is estimated as：543m.The safe speed estimated according to Class II highway truck appraising model in No. 2 bends is：82km/h.According to
Vehicle current driving speed and acceleration, the speed after estimating 10 seconds, early warning, otherwise not early warning are then carried out more than 82km/h.
The present embodiment on the basis of GPS real-time positioning informations and map datum, using extracted it is a range of before
Road in the rear node latitude and longitude coordinates calculate deflection difference, and are recognized and go off the curve or straight way with this, curved according to what is identified
Road, recycles curve fitting technique to realize the solution of bend equation, estimates the radius of curvature gone off the curve, and then estimates vehicle and leads to
The safe speed of bend is crossed, early warning is carried out based on this.Round-the-clock bend is realized on the premise of early warning accuracy is ensured
Identification and bend early warning, effectively increase the stability of bend early warning, it is ensured that traffic safety.
Embodiment six
Fig. 8 is refer to, the present embodiment is a kind of bend identifying system based on map datum of correspondence above-described embodiment, bag
Include：
Matching module 1, for matching vehicle location in map, obtains road where vehicle；
First obtains module 2, for obtaining the road section in default first distance in vehicle location rear on the road
Road circuit node in point and the second distance of vehicle location front preset, obtains node linked list；
First computing module 3, for calculating using each road circuit node as starting point in the node linked list, a road section below
Point is the deflection of the line segment of terminal；
Second computing module 4, the deflection difference for calculating the line segment and its previous line segment；
Mark module 5, the deflection of the road circuit node for the deflection difference to be designated as to the correspondence line segment starting point
Difference；
Second obtains module 6, for according to the deflection difference, obtaining bend node, obtaining bend chained list.
The matching module 1 includes：
First acquisition unit 101, the gps data current for obtaining vehicle；
Second acquisition unit 102, for the coordinate information in the gps data, obtains vehicle periphery preset range
Interior candidate roads, obtain the first candidate roads collection；
First judging unit 103, road and the connectedness of the candidate roads for judging the matching of vehicle last time；
Culling unit 104, rejects disconnected candidate roads for being concentrated in first candidate roads, obtains the second time
Seek road collection；
First computing unit 105, each candidate roads and the gps data are concentrated for calculating second candidate roads
In vector position information similarity；
Unit 106 is chosen, for choosing the minimum candidate roads of the similarity as road where vehicle.
The system also includes：
3rd computing module 7, for calculating the distance between two nodes adjacent in the node linked list；
Removing module 8, if being less than default 3rd distance for the distance, deletes any section in described two nodes
Point.
The system also includes：
First update module 9, if the deflection difference S ' for line segment described in one and its previous line segment is more than 180 degree,
Updating the deflection difference is | S ' | -360；
Second update module 10, if the deflection difference S ' for line segment described in one and its previous line segment is less than -180 degree,
The deflection difference is then updated for 360- | S ' |.
Described second, which obtains module 6, includes：
First division unit 601, for being, on the occasion of, negative value or zero situation, sequentially to divide described according to deflection difference
Road circuit node outside node linked list Zhong Chushou roads circuit node, obtains the node packet of sequential；
First adds unit 602, and first road circuit node for the node to be grouped is added to previous node packet
In；
First deletes unit 603, and the deflection difference for deleting remaining road circuit node in addition to extreme trace circuit node is zero
Node packet；
First obtains unit 604, and for obtaining being grouped the bend chained list constituted by remaining node, node described in one is grouped
One bend of correspondence；
Second adds unit 605, if for node packet only comprising two road circuit nodes, will be with described two roads
The adjacent previous road circuit node of circuit node is added in the node packet；
Traversal Unit 606, the bend node for traveling through each bend in the bend chained list；
Second computing unit 607, if being less than default first difference for the corresponding deflection difference of a bend node,
Calculate the distance of the bend node and next bend node；
Second division unit 608, if being more than default 4th distance for the distance, according to the bend node,
The corresponding bend of the bend node is divided into two bends；
3rd computing unit 609, the discrepancy angle absolute difference for calculating each bend respectively；
Second deletes unit 610, if the discrepancy angle absolute difference for a bend is less than default second difference,
Then the bend is deleted from bend chained list.
The system also includes：
Acquisition module 11, if being in a corresponding bend in the bend chained list after preset time for vehicle, is obtained
The corresponding bend set of node of the bend；
4th computing module 12, for according to the bend set of node, carrying out curve fitting, calculates the curvature of the bend
Radius.
4th computing module 12 includes：
Projecting cell 1201, for the bend node of the bend to be projected in rectangular coordinate system；
Rotary unit 1202, rotates to parallel institute for the line direction by the bend node according to bend first and last node
State the X-direction of rectangular coordinate system；
Translation unit 1203, the original for the bend node to be moved to the rectangular coordinate system according to bend first node
Point；
4th computing unit 1204, for calculating relative coordinate of each bend node with respect to bend first node；
Fitting unit 1205, for according to the relative coordinate, a quadratic polynomial most young waiter in a wineshop or an inn to be carried out to the bend node
Multiply fitting, obtain the fitting coefficient of curve；
5th computing unit 1206, for according to the fitting coefficient, calculating the radius of curvature of the bend.
In summary, the present invention is provided a kind of bend recognition methods and its system based on map datum, with map number
Judge according to combining, also can be real when the night of rainy day, insufficient light even in low visibility or pavement of road do not have any feature
When identify the degree of crook of vehicle front bend so that bend identification can not carried out with round-the-clock by road surface effect；Together
When, only obtain a range of road node data in vehicle location front and back to carry out bend identification, in the identification of real-time bend
While also ensure that efficiency；Bend set of node in bend chained list carries out curve fitting, and calculates the radius of curvature of bend,
Safe early warning is carried out available for bend speed, travel safety is improved.
Embodiments of the invention are the foregoing is only, are not intended to limit the scope of the invention, it is every to utilize this hair
The equivalents that bright specification and accompanying drawing content are made, or the technical field of correlation is directly or indirectly used in, similarly include
In the scope of patent protection of the present invention.
Claims (10)
1. a kind of bend recognition methods based on map datum, it is characterised in that including：
Vehicle location is matched in map, road where vehicle is obtained；
Obtain road circuit node and vehicle location front preset of the vehicle location rear on the road default first in
Road circuit node in second distance, obtains node linked list；
Calculate using each road circuit node as starting point in the node linked list, one of circuit node is the direction of the line segment of terminal below
Angle；
Calculate the deflection difference of the line segment and its previous line segment；
The deflection difference is designated as to the deflection difference of the road circuit node of the correspondence line segment starting point；
According to the deflection difference, bend node is obtained, bend chained list is obtained.
2. the bend recognition methods according to claim 1 based on map datum, it is characterised in that described " in map
Vehicle location is matched, road where vehicle is obtained " be specially：
Obtain the current gps data of vehicle；
According to the coordinate information in the gps data, the candidate roads in vehicle periphery preset range are obtained, the first candidate is obtained
Road collection；
Judge the road and the connectedness of the candidate roads of vehicle last time matching；
Concentrated in first candidate roads and reject disconnected candidate roads, obtain the second candidate roads collection；
Calculating second candidate roads concentrates each candidate roads similar to the vector position information in the gps data
Degree；
The minimum candidate roads of the similarity are chosen as road where vehicle.
3. the bend recognition methods according to claim 1 based on map datum, it is characterised in that described " to obtain node
After chained list ", further comprise：
Calculate the distance between two nodes adjacent in the node linked list；
If the distance is less than default 3rd distance, any node in described two nodes is deleted.
4. the bend recognition methods according to claim 1 based on the map datum, it is characterised in that " line segment
With the deflection difference of its previous line segment " after, further comprise：
If the deflection difference S ' of line segment described in one and its previous line segment is more than 180 degree, updating the deflection difference is | S '
|-360；
If the deflection difference S ' of line segment described in one and its previous line segment is less than -180 degree, updating the deflection difference is
360-|S’|。
5. the bend recognition methods according to claim 1 based on map datum, it is characterised in that described " according to described
Deflection difference, obtains bend node, obtains bend chained list " be specially：
It is, on the occasion of, negative value or zero situation, sequentially to divide outside the node linked list Zhong Chushou roads circuit node according to deflection difference
Road circuit node, obtain sequential node packet；
First road circuit node that the node is grouped is added in previous node packet；
Delete the node packet that the deflection difference of remaining road circuit node in addition to extreme trace circuit node is zero；
Obtain being grouped the bend chained list constituted by remaining node, node packet one bend of correspondence described in one.
6. the bend recognition methods according to claim 5 based on map datum, it is characterised in that described " to obtain by remaining
After the bend chained list that remaining node packet is constituted, one bend of node packet correspondence described in one ", further comprise：
If the node packet is only comprising two road circuit nodes, by the previous road circuit node adjacent with described two road circuit nodes
It is added in the node packet.
7. the bend recognition methods according to claim 5 based on map datum, it is characterised in that described " to obtain by remaining
After the bend chained list that remaining node packet is constituted, one bend of node packet correspondence described in one ", further comprise：
Travel through the bend node of each bend in the bend chained list；
If the corresponding deflection difference of a bend node be less than default first difference, calculate the bend node with it is next curved
The distance of road node；
If the distance is more than default 4th distance, according to the bend node, by the corresponding bend of the bend node
It is divided into two bends；
The discrepancy angle absolute difference of each bend is calculated respectively；
If the discrepancy angle absolute difference of a bend is less than default second difference, by the bend from bend chained list
Delete.
8. the bend recognition methods according to claim 1 based on map datum, it is characterised in that described " according to described
Deflection difference, obtains bend node, obtains bend chained list " after, further comprise：
If vehicle is in a corresponding bend in the bend chained list after preset time, the corresponding bend section of the bend is obtained
Point set；
According to the bend set of node, carry out curve fitting, calculate the radius of curvature of the bend.
9. the bend recognition methods according to claim 8 based on map datum, it is characterised in that described " according to described
Bend set of node, carries out curve fitting, and calculates the radius of curvature of the bend " be specially：
The bend node of the bend is projected in rectangular coordinate system；
Line direction by the bend node according to bend first and last node is rotated to the X-axis side of the parallel rectangular coordinate system
To；
The bend node is moved to the origin of the rectangular coordinate system according to bend first node；
Calculate relative coordinate of each bend node with respect to bend first node；
According to the relative coordinate, quadratic polynomial least square fitting is carried out to the bend node, the fitting of curve is obtained
Coefficient；
According to the fitting coefficient, the radius of curvature of the bend is calculated.
10. a kind of bend identifying system based on map datum, it is characterised in that including：
Matching module, for matching vehicle location in map, obtains road where vehicle；
First obtains module, for obtaining road circuit node and car in default first distance in vehicle location rear on the road
Road circuit node in the second distance of position front preset, obtains node linked list；
First computing module, for calculating using each road circuit node as starting point in the node linked list, one of circuit node is below
The deflection of the line segment of terminal；
Second computing module, the deflection difference for calculating the line segment and its previous line segment；
Mark module, the deflection difference of the road circuit node for the deflection difference to be designated as to the correspondence line segment starting point；
Second obtains module, for according to the deflection difference, obtaining bend node, obtaining bend chained list.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610277789.4A CN107328423B (en) | 2016-04-28 | 2016-04-28 | Curve identification method and system based on map data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610277789.4A CN107328423B (en) | 2016-04-28 | 2016-04-28 | Curve identification method and system based on map data |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107328423A true CN107328423A (en) | 2017-11-07 |
| CN107328423B CN107328423B (en) | 2020-10-16 |
Family
ID=60193122
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610277789.4A Active CN107328423B (en) | 2016-04-28 | 2016-04-28 | Curve identification method and system based on map data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107328423B (en) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108674416A (en) * | 2018-03-21 | 2018-10-19 | 厦门雅迅网络股份有限公司 | A method of it is broadcast based on vehicle bus and rebuilds curve ahead information |
| CN109033176A (en) * | 2018-06-25 | 2018-12-18 | 腾讯科技（深圳）有限公司 | Determination method, apparatus, storage medium and the computer equipment of road curvature |
| CN109584570A (en) * | 2018-12-29 | 2019-04-05 | 浙江方大智控科技有限公司 | Traffic management method based on roadside fixed test equipment |
| CN111611901A (en) * | 2020-05-15 | 2020-09-01 | 北京百度网讯科技有限公司 | Vehicle reverse running detection method, device, equipment and storage medium |
| CN111780773A (en) * | 2019-04-04 | 2020-10-16 | 北京嘀嘀无限科技发展有限公司 | Method and system for identifying curve |
| CN112050820A (en) * | 2020-09-02 | 2020-12-08 | 平安科技（深圳）有限公司 | Road matching method and device, electronic equipment and readable storage medium |
| CN112686904A (en) * | 2020-12-14 | 2021-04-20 | 深兰人工智能(深圳)有限公司 | Lane division method, lane division device, electronic equipment and storage medium |
| US11420649B2 (en) * | 2020-03-24 | 2022-08-23 | Here Global B.V. | Method, apparatus, and computer program product for generating turn paths through an intersection |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP1086844A2 (en) * | 1999-09-21 | 2001-03-28 | Fuji Jukogyo Kabushiki Kaisha | Road shape estimation apparatus and curve approach control apparatus |
| CN1648605A (en) * | 2004-01-30 | 2005-08-03 | 爱信艾达株式会社 | Apparatus for predicting road shape |
| JP2006004029A (en) * | 2004-06-16 | 2006-01-05 | Sumitomo Denko Field System Kk | Device for detecting curve section |
| CN1788421A (en) * | 2003-10-17 | 2006-06-14 | 松下电器产业株式会社 | Encoding data generation method and device |
| CN102568009A (en) * | 2010-12-17 | 2012-07-11 | 上海博泰悦臻电子设备制造有限公司 | Line segment reducing device and method for electronic map |
| CN103310699A (en) * | 2012-03-16 | 2013-09-18 | 北京四维图新科技股份有限公司 | Method for extracting road alignment parameters |
| CN103617731A (en) * | 2013-09-09 | 2014-03-05 | 重庆大学 | Method for generating road network vector map utilizing GPS data of floating vehicles in city |
| CN104061936A (en) * | 2013-03-19 | 2014-09-24 | 福特全球技术公司 | Method Of Building And Using Local Map Of Vehicle Drive Path |
-
2016
- 2016-04-28 CN CN201610277789.4A patent/CN107328423B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP1086844A2 (en) * | 1999-09-21 | 2001-03-28 | Fuji Jukogyo Kabushiki Kaisha | Road shape estimation apparatus and curve approach control apparatus |
| CN1788421A (en) * | 2003-10-17 | 2006-06-14 | 松下电器产业株式会社 | Encoding data generation method and device |
| CN1648605A (en) * | 2004-01-30 | 2005-08-03 | 爱信艾达株式会社 | Apparatus for predicting road shape |
| JP2006004029A (en) * | 2004-06-16 | 2006-01-05 | Sumitomo Denko Field System Kk | Device for detecting curve section |
| CN102568009A (en) * | 2010-12-17 | 2012-07-11 | 上海博泰悦臻电子设备制造有限公司 | Line segment reducing device and method for electronic map |
| CN103310699A (en) * | 2012-03-16 | 2013-09-18 | 北京四维图新科技股份有限公司 | Method for extracting road alignment parameters |
| CN104061936A (en) * | 2013-03-19 | 2014-09-24 | 福特全球技术公司 | Method Of Building And Using Local Map Of Vehicle Drive Path |
| CN103617731A (en) * | 2013-09-09 | 2014-03-05 | 重庆大学 | Method for generating road network vector map utilizing GPS data of floating vehicles in city |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108674416A (en) * | 2018-03-21 | 2018-10-19 | 厦门雅迅网络股份有限公司 | A method of it is broadcast based on vehicle bus and rebuilds curve ahead information |
| CN109033176A (en) * | 2018-06-25 | 2018-12-18 | 腾讯科技（深圳）有限公司 | Determination method, apparatus, storage medium and the computer equipment of road curvature |
| CN109033176B (en) * | 2018-06-25 | 2024-02-27 | 腾讯科技（北京）有限公司 | Road curvature determination method, device, storage medium and computer equipment |
| CN109584570A (en) * | 2018-12-29 | 2019-04-05 | 浙江方大智控科技有限公司 | Traffic management method based on roadside fixed test equipment |
| CN111780773A (en) * | 2019-04-04 | 2020-10-16 | 北京嘀嘀无限科技发展有限公司 | Method and system for identifying curve |
| US11420649B2 (en) * | 2020-03-24 | 2022-08-23 | Here Global B.V. | Method, apparatus, and computer program product for generating turn paths through an intersection |
| CN111611901A (en) * | 2020-05-15 | 2020-09-01 | 北京百度网讯科技有限公司 | Vehicle reverse running detection method, device, equipment and storage medium |
| CN111611901B (en) * | 2020-05-15 | 2023-10-03 | 北京百度网讯科技有限公司 | Vehicle reverse running detection method, device, equipment and storage medium |
| CN112050820A (en) * | 2020-09-02 | 2020-12-08 | 平安科技（深圳）有限公司 | Road matching method and device, electronic equipment and readable storage medium |
| CN112050820B (en) * | 2020-09-02 | 2024-05-07 | 平安科技（深圳）有限公司 | Road matching method, device, electronic equipment and readable storage medium |
| CN112686904A (en) * | 2020-12-14 | 2021-04-20 | 深兰人工智能(深圳)有限公司 | Lane division method, lane division device, electronic equipment and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107328423B (en) | 2020-10-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107328423A (en) | Bend recognition methods and its system based on map datum | |
| CN111912416B (en) | Method, device and equipment for positioning equipment | |
| CN106767853B (en) | Unmanned vehicle high-precision positioning method based on multi-information fusion | |
| CN110146910B (en) | Positioning method and device based on data fusion of GPS and laser radar | |
| CN102208035B (en) | Image processing system and position measuring system | |
| Tao et al. | Lane marking aided vehicle localization | |
| CN108388641B (en) | Traffic facility map generation method and system based on deep learning | |
| CN108171131B (en) | Improved MeanShift-based method for extracting Lidar point cloud data road marking line | |
| CN104089619B (en) | The accurate matching system of GPS navigation map of pilotless automobile and operational approach thereof | |
| CN112904395B (en) | Mining vehicle positioning system and method | |
| EP2113746A1 (en) | Feature information collecting device and feature information collecting program, and vehicle position recognizing device and navigation device | |
| CN110332945B (en) | Vehicle navigation method and device based on traffic road marking visual identification | |
| Wang et al. | Vehicle localization at an intersection using a traffic light map | |
| JP2015004814A (en) | Lane map generation device and program | |
| EP3093620B1 (en) | System and method for detecting roundabouts from probe data using vector fields | |
| CN113358125B (en) | Navigation method and system based on environment target detection and environment target map | |
| CN109325390B (en) | Positioning method and system based on combination of map and multi-sensor detection | |
| CN110431609A (en) | Vehicle location estimating device | |
| CN112082567A (en) | Map path planning method based on combination of improved Astar and Grey wolf algorithm | |
| CN114565674B (en) | Method and device for purely visually positioning urban structured scene of automatic driving vehicle | |
| CN115564865A (en) | Construction method and system of crowdsourcing high-precision map, electronic equipment and vehicle | |
| Zinoune et al. | Detection of missing roundabouts in maps for driving assistance systems | |
| CN111126327B (en) | Lane line detection method and system, vehicle-mounted system and vehicle | |
| JPH0727541A (en) | Measuring device for road shape and vehicle position | |
| CN114889606A (en) | Low-cost high-precision positioning method based on multi-sensor fusion |
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 | ||
| CP03 | Change of name, title or address | ||
| CP03 | Change of name, title or address |
Address after: 303-e, Zone C, innovation building, software park, torch hi tech Zone, Xiamen City, Fujian Province Patentee after: Xiamen Yaxun Zhilian Technology Co.,Ltd. Country or region after: China Address before: Yaxun building, No.46, guanri Road, phase II, software park, Xiamen, Fujian, 361000 Patentee before: XIAMEN YAXON NETWORK Co.,Ltd. Country or region before: China |