CN108106619A - Main and side road recognition methods and its system - Google Patents
Main and side road recognition methods and its system Download PDFInfo
- Publication number
- CN108106619A CN108106619A CN201611058670.4A CN201611058670A CN108106619A CN 108106619 A CN108106619 A CN 108106619A CN 201611058670 A CN201611058670 A CN 201611058670A CN 108106619 A CN108106619 A CN 108106619A
- Authority
- CN
- China
- Prior art keywords
- main
- path adaptation
- course angle
- membership
- window phase
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
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
-
- 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
- G01C21/34—Route searching; Route guidance
- G01C21/3446—Details of route searching algorithms, e.g. Dijkstra, A*, arc-flags, using precalculated routes
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S19/00—Satellite radio beacon positioning systems; Determining position, velocity or attitude using signals transmitted by such systems
- G01S19/38—Determining a navigation solution using signals transmitted by a satellite radio beacon positioning system
- G01S19/39—Determining a navigation solution using signals transmitted by a satellite radio beacon positioning system the satellite radio beacon positioning system transmitting time-stamped messages, e.g. GPS [Global Positioning System], GLONASS [Global Orbiting Navigation Satellite System] or GALILEO
- G01S19/42—Determining position
- G01S19/45—Determining position by combining measurements of signals from the satellite radio beacon positioning system with a supplementary measurement
- G01S19/47—Determining position by combining measurements of signals from the satellite radio beacon positioning system with a supplementary measurement the supplementary measurement being an inertial measurement, e.g. tightly coupled inertial
Landscapes
- Engineering & Computer Science (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Automation & Control Theory (AREA)
- Computer Networks & Wireless Communication (AREA)
- Navigation (AREA)
Abstract
The present invention, which provides main and side road recognition methods and its system, method, to be included：When the path adaptation similarity maximum difference of adjacent GPS point in the current GPS point and the path adaptation similarity of matched road that vehicle enters window phase, and foundation obtains in real time, acquisition window phase；The course angle accumulated change value in yaw velocity value change sequence acquisition window phase obtained according to gyroscope；According to default membership function, the degree of membership of the path adaptation similarity maximum difference and course angle accumulated change value is calculated respectively；Calculating acquisition again includes straight-going mode and the Comprehensis pertaining of lane change pattern.The present invention can overcome the dependence in recognition result focus ring border；It can exclude the influence that traffic behavior brings the influence of speed factor to recognition result；Bend environment can be preferably suitable for；Two independent input parameters have been merged, while promoting accuracy of identification, and can guarantee the influence brought during the failure of one parameter.
Description
Technical field
The present invention relates to vehicle mounted guidance fields, particularly relate to main and side road recognition methods and its system.
Background technology
In vehicle mounted guidance field, terminal needs the road network to map datum according to the coordinate real-time matching that sensor obtains
On, and between some parallel main and side roads being closer to, due to the limitation of vehicle GPS precision, often cannot all reach fine
Effect.
Some traditional technologies are in order to solve the problems, such as that this comes into being：Patent《A kind of on-vehicle navigation apparatus and identification master
The air navigation aid of bypass》With《The matching process of the road main and side road of navigator》Combine vehicle-mounted camera and GPS sensor, energy
It is enough more accurately to identify main and side road with assisting navigation, but since camera is susceptible to the influence of ambient light environment, do not following the Way
Road car diatom or it can not play effect in the environment of rainy day, the relatively low insufficient light of night visibility；Patent《A kind of movement
The main and side road recognition methods of terminal and device》Speed measuring device identification main and side road is applied, but in traffic congestion, the friendship having some setbacks of driving a vehicle
When under logical state, the identification of mistake can be also generated according to speed；Paper《Entrance vehicle behavior pattern in through street in map match
Identification》The mode for employing Fuzzy Pattern Recognition combines course angle and speed factor, but the traffic shape equally having some setbacks in driving
When under state or in the main and side road with bend, course angle and speed also can be disturbed greatly, it is difficult in bend and complexity
It realizes and accurately identifies under traffic behavior environment.
Therefore, it is necessary to a kind of main and side road recognition methods and system are provided, to solve the above problems well.
This technology will solve above-mentioned technology problem encountered.This technology has used for reference paper《Through street goes out in map match
Entrance vehicle behavior pattern-recognition》The thought of middle Fuzzy Pattern Recognition combines gyroscope and GPS data, but not by speed
Factor brings main and side road basis of characterization into.Sensitivity of the camera for luminous environment so is overcome, and eliminates traffic
Influence of the situation to speed factor.This technology also optimizes influence of the bend for gyroscope simultaneously, the road that main and side road is identified
Road environment has been expanded to bend.
The content of the invention
The technical problems to be solved by the invention are：Main and side road recognition methods and system are provided, overcome camera for light
The sensitivity of environment excludes the influence of traffic, improves main and side road recognition accuracy, has expanded recognizable road environment.
In order to solve the above-mentioned technical problem, the technical solution adopted by the present invention is：
Main and side road recognition methods, including：
According to historical empirical data, course angle accumulated change value and the person in servitude of path adaptation similarity sequence maximum difference are preset
Category degree function；
By vehicle match to map datum road network；
When vehicle enters window phase, the window phase is what is drawn a circle to approve centered on the nearest main and side road bifurcation of vehicle front
One scope；
According to the current GPS point obtained in real time and the path adaptation similarity of matched road, obtain adjacent in window phase
The path adaptation similarity maximum difference of GPS point；
The course angle accumulated change value in yaw velocity value change sequence acquisition window phase obtained according to gyroscope；
According to the membership function, the path adaptation similarity maximum difference and course angle accumulated change are calculated respectively
The degree of membership of value；
According to the degree of membership, calculating includes straight-going mode and the Comprehensis pertaining of lane change pattern.
Another technical solution provided by the invention is：
Main and side road identifying system, including：
Presetting module, for according to historical empirical data, presetting course angle accumulated change value and path adaptation similarity sequence
The membership function of row maximum difference；
Matching module, for by vehicle match to map datum road network；
First acquisition module enters window phase for working as vehicle, and the window phase is with the nearest main and side road of vehicle front
The scope drawn a circle to approve centered on bifurcation；The current GPS point and the path adaptation similarity of matched road that foundation obtains in real time,
Obtain the path adaptation similarity maximum difference of adjacent GPS point in window phase；
Second acquisition module, the course angle in yaw velocity value change sequence acquisition window phase obtained according to gyroscope
Accumulated change value；
First computing module, for according to the membership function, it is maximum poor to calculate the path adaptation similarity respectively
The degree of membership of value and course angle accumulated change value；
Second computing module, for according to the degree of membership, calculating includes straight-going mode and the synthesis of lane change pattern is subordinate to
Degree.
The beneficial effects of the present invention are：The present invention relies only on gyroscope and GPS sensor, using electronic map data and
Map matching technology can accurately identify vehicle when close to main and side road connection road, if there is lane change switching action, so that it is determined that
Whether vehicle is still being travelled on main road/bypass after connecting road.The present invention knows not against traditional camera image
Other effect does not depend on luminous environment, overcomes camera focus ring border sensitive issue；Meanwhile Comprehensis pertaining has been merged according to top
For the data that spiral shell instrument and GPS sensor obtain respectively as input parameter, two input parameters are mutual indepedent, are independent of each other, in time
One failure nor affects on recognition result, the accuracy of recognition result.
Description of the drawings
Fig. 1 is the flow diagram of the main and side road recognition methods of the present invention；
Fig. 2 is schematic diagram of the window phase in map datum in the present invention；
Fig. 3 is the schematic diagram of fitting circle in bend compensation process of the present invention；
Fig. 4 is the flow diagram of the embodiment of the present invention one；
Fig. 5 is the structure composition schematic diagram of the main and side road identifying system of the present invention；
Fig. 6 is the structure composition schematic diagram of the embodiment of the present invention two.
Label declaration：
1st, presetting module；2nd, matching module；3rd, the first acquisition module；4th, the second acquisition module；
5th, the first computing module；6th, the second computing module；
41st, bend compensating unit；42nd, filter unit；43rd, judging unit；
51st, structural unit；52nd, computing unit；
61st, unit is preset；62nd, analytic unit.
Specific embodiment
For the technology contents that the present invention will be described in detail, the objects and the effects, below in conjunction with embodiment and coordinate attached
Figure is explained.
The design of most critical of the present invention is：Only in accordance with the data that gyroscope and GPS sensor obtain, accurately identify in vehicle
By main and side road connection road when whether lane change.
Explanation of technical terms of the present invention:
Referring to Fig.1 and 2, the present invention provides a kind of main and side road recognition methods, including：
According to historical empirical data, course angle accumulated change value and the person in servitude of path adaptation similarity sequence maximum difference are preset
Category degree function；
By vehicle match to map datum road network；
When vehicle enters window phase, the window phase is what is drawn a circle to approve centered on the nearest main and side road bifurcation of vehicle front
One scope；
According to the current GPS point obtained in real time and the path adaptation similarity of matched road, obtain adjacent in window phase
The path adaptation similarity maximum difference of GPS point；
The course angle accumulated change value in yaw velocity value change sequence acquisition window phase obtained according to gyroscope；
According to the membership function, the path adaptation similarity maximum difference and course angle accumulated change are calculated respectively
The degree of membership of value；
According to the degree of membership, calculating includes straight-going mode and the Comprehensis pertaining of lane change pattern.
Further, the yaw velocity value change sequence obtained according to gyroscope obtains the course angle in window phase
Accumulated change value, specially：
Bend compensation is carried out to the yaw velocity value sequence of gyroscope output, according to the yaw velocity value after compensation
The integral result of sequence judges whether lane-changing intention.
If being straight way in window phase, offset 0.
Seen from the above description, it can identify main and side road in corner, expand the scope of application of main and side road identification.
Further, the result that foundation integrates the yaw velocity value sequence after compensation judges whether that lane-change is anticipated
Figure, specially：
Mean filter processing is carried out to the yaw velocity value sequence after integration, according to filtered yaw velocity value sequence
Row integral result judges whether lane-changing intention.
Seen from the above description, can filter out that vehicle-mounted gyroscope obtains in vehicle travel process is mingled with noise, carries
High identification accuracy.
Further, it is described according to the membership function, calculate respectively the path adaptation similarity maximum difference and
The degree of membership of course angle accumulated change value, specially：
Construct the set of modes of straight-going mode and lane change pattern composition；
Using the course angle accumulated change value and path adaptation similarity maximum difference as input construction feature vector, calculate
The degree of membership of each input construction feature vector.
Further, described according to the degree of membership, calculating includes straight-going mode and the Comprehensis pertaining of lane change pattern, has
Body is：
It is worth corresponding weight according to the default path adaptation similarity maximum difference and course angle accumulated change
Value, calculating include straight-going mode and the Comprehensis pertaining of lane change pattern；
The Comprehensis pertaining is analyzed according to maximum subjection principle, determines vehicle in window phase in straight trip mould
Formula or lane change pattern.
Referring to Fig. 5, another technical solution provided by the invention is：
Main and side road identifying system, including：
Presetting module, for according to historical empirical data, presetting course angle accumulated change value and path adaptation similarity sequence
The membership function of row maximum difference；
Matching module, for by vehicle match to map datum road network；
First acquisition module enters window phase for working as vehicle, and the window phase is with the nearest main and side road of vehicle front
The scope drawn a circle to approve centered on bifurcation；The current GPS point and the path adaptation similarity of matched road that foundation obtains in real time,
Obtain the path adaptation similarity maximum difference of adjacent GPS point in window phase；
Second acquisition module, for the acquisition boat in the yaw velocity value change sequence window phase according to gyroscope acquisition
To angle accumulated change value；
First computing module, for according to the membership function, it is maximum poor to calculate the path adaptation similarity respectively
The degree of membership of value and course angle accumulated change value；
Second computing module, for according to the degree of membership, calculating includes straight-going mode and the synthesis of lane change pattern is subordinate to
Degree.
Further, second acquisition module includes：
Bend compensating unit, the yaw velocity value sequence for being exported to gyroscope carry out bend compensation；
Judging unit judges whether lane-change for foundation to the integral result of the yaw velocity value sequence after compensation
It is intended to.
Further, second acquisition module further includes filter unit, for the yaw velocity value sequence after integration
Row carry out mean filter processing；
The judging unit, specifically for being judged whether according to filtered yaw velocity value sequence integral result
Lane-changing intention.
Further, first computing module includes：
Structural unit, for constructing the set of modes of straight-going mode and lane change pattern composition；
Computing unit, for being constructed using the course angle accumulated change value and path adaptation similarity maximum difference as input
Feature vector calculates the degree of membership of each input construction feature vector.
Further, second computing module includes：
Default unit, for each according to the default path adaptation similarity maximum difference and course angle accumulated change value
Self-corresponding weighted value, calculating include straight-going mode and the Comprehensis pertaining of lane change pattern；
Analytic unit for being analyzed according to maximum subjection principle the Comprehensis pertaining, determines vehicle in window
Straight-going mode or lane change pattern are in phase.
Embodiment one
Fig. 2 to Fig. 4 is refer to, the present embodiment provides a kind of main and side roads to know method for distinguishing, can exist accurately identifying vehicle
When connecting road by main and side road, if having switching action, so as to which indirect discrimination goes out whether vehicle is still going after connecting road
It sails on original path.
Specifically, including：
(1) preliminary preparation：
(1) according to empirical data progress statistical analysis of driving a vehicle to substantial amounts of history as a result, default course angle accumulated change
Value and the membership function of path adaptation similarity sequence maximum difference；
(2) it is the scope drawn a circle to approve centered on the nearest main and side road bifurcation of vehicle front to define window phase, i.e., with vehicle
Centered on exercising the nearest main and side road connection road first point in front, (such as 170 meters of front, 120 meters of rear) in a certain range
The vehicle road to be passed through.The first point is road and the bifurcation being connected, and by taking Fig. 2 as an example, Fig. 2 is real world Road
The mapable of net shows that the thicker line of two of which (between 4 points to 5 points and) represents the connection road of main road and bypass at 8 points,
The node of road is formed in point position electronic map data residing for label 1-8；What " first point " referred to be exactly in figure each road with
Connect the place of road fork, i.e. 4 points and 8 points in Fig. 2, it should be noted that：6 points are not first points because it for point and
Non- bifurcation point.
(2) data acquisition phase
Data acquisition phase, mainly conversion and the data in the acquisition window phase, the data of acquisition include vehicle in window
When in the phase, the yaw velocity value sequence by bend compensation that gyroscope obtains is integrated, according to the sequence after integration
With the path adaptation similarity obtained after map match obtained by GPS sensor.Specifically, data acquisition phase can
To comprise the following steps：
(1) by vehicle match to map datum road network
Preferably, the location information that can be exported according to GPS sensor, passes through the map match skill of traditional local optimum
Art exercises vehicle match on road to current, and records the similarity with institute matched road, i.e., current GPS point and
Path adaptation similarity with road.
The map-matching algorithm of local optimum can be summarized as：Determine a match point every time, next point is from having determined
Match point starts, and finds local optimum side according to distance and directional similarity every time.It is by the similarity of road and GPS point position
It is defined as：
S=waa+wdd；
Wherein a is GPS point position course angle and road azimuth angle difference, waFor the weighted value of angle difference part；D is
The most short Euclidean distance in GPS point Wei Dao roads roadside, wdFor the weighted value apart from part, wd+wa=1.
(2) the search window phase
Judge that vehicle whether into window phase, if it is determined that into window phase, then performs (3) and (4) simultaneously.
(3) the path adaptation similarity of each GPS point in window phase is obtained
Path adaptation similarity of the current GPS point with matched road is obtained in real time, i.e., each GPS point position and institute in window
Match the path adaptation similarity of road.
(4) bend compensation is carried out
The output valve (result of yaw velocity value sequence integration) of gyroscope z-axis is present with apparent change in Ackermann steer angle
Change, but the turning that the turn inside diameter in window phase may be carried out not for switching main and side road, when vehicle is travelled in bend
It can be forced to turn due to Road turnings, bend compensation seeks to compensate this part of gyroscope z-axis variable quantity, makes gyroscope
The influence of racetrack portion is subtracted out in z-axis variable quantity.Certainly, if straight way, bend offset is just zero.
The step of bend compensates can be divided into：
4.1 circuit node fitting circles；
It is compared due to electronic map data road circuit node with the newer tracing points of GPS and wants much sparse.Therefore, can utilize
Map datum road circuit node, using three nodes as one group of fitting circle.As shown in figure 3, node 2,3 and 4 fits round R1, node 3,
4 and 5 fit round R2；Between point 4 and point 5 on the right side of connecting line using point 4 for exit line as main road and bypass company
It connects；
4.2 calculate grid azimuth of the GPS point position in fitting circle；
The newer frequencies of GPS can ensure at least one GPS point position between two road circuit nodes.Therefore it is two neighboring
GPS point position can calculate the grid azimuth (y-axis angle clockwise) of the subpoint on circle in same fitting circle.
As shown in figure 3, each comfortable fitting circle R of GPS point position A and B can be calculated1The grid azimuth α of upper subpoint and
β, GPS point position B and C are in fitting circle R2On grid azimuth γ and θ.
4.3 calculate two neighboring GPS point position grid azimuth difference；
As shown in figure 3, the grid azimuth difference β-α of GPS point position A and B can be calculated；The coordinate of GPS point position B and C
Azimuth difference θ-γ.
4.4 compensate gyro data using adjacent two point coordinates azimuth difference；
Using the z-axis data of gyroscope, the angle that is turned between the two adjacent GPS point positions that can add up out.Pass through 4.3 meters
The grid azimuth difference of calculating can compensate the cumulative angle.As shown in Figure 3, it is assumed that gyroscope between GPS point position A and B
Cumulative angle is G1, and gyroscope adds up angle as G2 between B and C, then the gyro data after compensating is respectively G1- (β-α),
G2-(θ-γ)。
(3) data analysis phase
Data analysis phase is divided into two main lines, what the data and GPS matching results that analysis gyroscope obtains respectively generated
Path adaptation similarity.Specifically, it may comprise steps of：
(1) gyro data sequence is filtered
Since vehicle-mounted gyroscope can be mingled with noise in vehicle travel process, it is therefore desirable to which the gyroscope after integration is obtained
Yaw velocity value sequence be filtered operation, the key step of filter operation can be divided into again：
1.1 sliding window mean filters；
It is that a window carries out mean filter, this mistake to yaw velocity value sequence with a default period, such as 3s
Journey mainly filters out sensor noise.
1.2 using zero axle by sequences segmentation as various pieces；
1.3 delete the part after window phase head, tail segmentation；
The window searched using data acquisition phase can block some continuous waveforms, window phase end to end if there is one
A little nonzero values, then for window phase outside continuity of the waveform in window phase, this part need to cast out with exclusive PCR.
1.4, which delete two adjacent positive negative part absolute values, mutually sums it up the part after the segmentation less than certain threshold value；
The lane change process waveform that gyroscope z-axis is recorded, which is similar to, has positive and negative bimodal sine curve, and due to usual
Main and side road connection road has certain length, will not be restored to original enforcement direction after Vehicular turn at once, therefore vehicle switches
The wave character of main and side road is then a unimodal curve, this process mainly filters out the lane change of non-switching intention in window phase
It influences.
Part after 1.5 surplus divisions is pressed sorts with the distance of window phase central point, finally chooses nearest from central point
That part.
(2) course angle accumulated change value is calculated
The part obtained after (1) is filtered is after having the yaw velocity value integration of switching intention in window phase
Sequence, i.e. course angle accumulated change sequence；Accumulation summation is carried out to this partial sequence, obtains the presence of switching lane change meaning in window phase
Course angle accumulated change value during figure.
(3) path adaptation similarity maximum difference is calculated
According to each GPS point position that data acquisition phase (3) step obtains and the path adaptation similarity group of matched road
Into sequence, calculate and obtain its maximum difference between any two, i.e., the path adaptation similarity maximum of adjacent GPS point in window phase
Difference.
(4) course angle accumulated change value and the degree of membership of path adaptation similarity maximum difference are calculated respectively
According to the course angle accumulated change value and path adaptation similarity sequence maximum difference constructed in preliminary preparation
The person in servitude for calculating course angle accumulated change value and path adaptation similarity maximum difference is calculated in respective membership function respectively
Category degree.Specific calculating process can include：
Set of modes R={ the R of 4.1 construction straight-going modes and switch mode composition1,R2}；Wherein, R represents set of modes,
R1 represents straight-going mode, and R2 represents lane change pattern；
4.2 add up changing value, path adaptation similarity maximum difference as input construction feature vector X=(v using course angled,
vs)；Wherein, X representative features vector, vdRepresent course angle accumulated change value of the vehicle in ring road window phase, vsRepresent that vehicle exists
The maximum difference of matching similarity in ring road window phase
4.3 calculate the degree of membership for obtaining each characteristic component.
(5) Comprehensis pertaining is calculated
What 4.3 in fusion previous step obtained in weighted fashion is subordinate to angle value, which can also be divided into two steps：
5.1 according to each characteristic component of setting weighted value wvd、wvsCalculate Comprehensis pertaining：
Wherein, wvd、wvsRespectively course angle accumulated change
The weighted value of value tag component and path adaptation similarity maximum difference characteristic component, and wvd+wvs=1
For the membership function of straight-going state course angle accumulated change component fuzzy set；
For the membership function of switching state course angle accumulated change component fuzzy set；
vdRepresent course angle accumulated change value of the vehicle in ring road window phase；
For the membership function of straight-going state path adaptation similarity maximum difference component fuzzy set；
R2vs(vs) ∈ [0,1] be switching state path adaptation similarity maximum difference component fuzzy set membership function；
vsRepresent the maximum difference of vehicle matching similarity in ring road window phase；
R1(v) Comprehensis pertaining of straight-going mode；
R2(v) Comprehensis pertaining of switch mode；
5.2 determine that vehicle is in straight-going mode or switch mode (lane change mould in window phase according to maximum subjection principle
Formula).
The present embodiment has the following advantages：1st, merely with gyro data and GPS sensor data, tradition is not utilized
Camera image, recognition effect do not depend on luminous environment；2nd, course angle changing value has been merged and path adaptation similarity is maximum poor
Two input parameters of value, the two amounts are gathered respectively by gyroscope and GPS sensor, and the two is mutual indepedent, is independent of each other, two
One of sensor, which fails entirety, can also export effective result；3rd, course angle changing value employs bend compensation so that this
Embodiment is also applied for the main and side road identification under bend environment.
Embodiment two
Fig. 6 is refer to, the present embodiment provides a kind of main and side road identifying system on the basis of embodiment one, including：
Presetting module 1, for according to historical empirical data, presetting course angle accumulated change value and path adaptation similarity sequence
The membership function of row maximum difference；
Matching module 2, for by vehicle match to map datum road network；
First acquisition module 3 enters window phase for working as vehicle, and the window phase is with the nearest main and side road of vehicle front
The scope drawn a circle to approve centered on bifurcation；The current GPS point and the path adaptation similarity of matched road that foundation obtains in real time,
Obtain the path adaptation similarity maximum difference of adjacent GPS point in window phase；
Second acquisition module 4, the acquisition course in yaw velocity value change sequence window phase obtained according to gyroscope
Angle accumulated change value；Specifically, second acquisition module includes bend compensating unit41, filter unit42And judging unit43；Bend compensating unit41Yaw velocity value sequence for being exported to gyroscope carries out bend compensation；Filter unit52, use
In to the yaw velocity value sequence progress mean filter processing after compensation；Judging unit43, for according to the horizontal stroke after compensation
The integral result of pivot angle speed value sequence judges whether lane-changing intention；
First computing module 5, for according to the membership function, it is maximum poor to calculate the path adaptation similarity respectively
The degree of membership of value and course angle accumulated change value；Specifically, the first computing module 5 includes structural unit 51 and computing unit 52,
Structural unit 51 is used to construct the set of modes of straight-going mode and lane change pattern composition；Computing unit 52, for the course
Angle accumulated change value and path adaptation similarity maximum difference are input construction feature vector, calculate each input construction feature vector
Degree of membership.
Second computing module 6, for according to the degree of membership, calculating includes straight-going mode and the synthesis of lane change pattern is subordinate to
Degree；Specifically, second computing module 6 includes default unit 61 and analytic unit 62；Default unit 61 is used for according to default
The path adaptation similarity maximum difference and course angle accumulated change be worth corresponding weighted value, calculating includes straight trip mould
The Comprehensis pertaining of formula and lane change pattern；Analytic unit 62, for being carried out according to maximum subjection principle to the Comprehensis pertaining
Analysis determines that vehicle is in straight-going mode or lane change pattern in window phase.
In conclusion main and side road recognition methods provided by the invention and its system, can not only overcome recognition result to light
The dependence of environment；And the influence that traffic behavior brings the influence of speed factor to recognition result can be excluded again；Further
, moreover it is possible to identification caused by overcome thing problems, such as bend is inaccurate, is preferably suitable for bend environment；Further, two have been merged
A independent input parameter while promoting accuracy of identification, and can guarantee the influence brought during the failure of one parameter.
The foregoing is merely the embodiment of the present invention, are not intended to limit the scope of the invention, every to utilize this hair
The equivalents that bright specification and accompanying drawing content are made directly or indirectly are used in relevant technical field, similarly include
In the scope of patent protection of the present invention.
Claims (10)
1. main and side road recognition methods, which is characterized in that including：
According to historical empirical data, course angle accumulated change value and the degree of membership of path adaptation similarity sequence maximum difference are preset
Function；
By vehicle match to map datum road network；
When vehicle enters window phase, the window phase is the model drawn a circle to approve centered on the nearest main and side road bifurcation of vehicle front
It encloses；
According to the current GPS point obtained in real time and the path adaptation similarity of matched road, adjacent GPS point in window phase is obtained
Path adaptation similarity maximum difference；
The course angle accumulated change value in yaw velocity value change sequence acquisition window phase obtained according to gyroscope；
According to the membership function, the path adaptation similarity maximum difference and course angle accumulated change value are calculated respectively
Degree of membership；
According to the degree of membership, calculating includes straight-going mode and the Comprehensis pertaining of lane change pattern.
2. main and side road recognition methods as described in claim 1, which is characterized in that the yaw angle speed obtained according to gyroscope
Angle value change sequence obtains the course angle accumulated change value in window phase, is specially：
Bend compensation is carried out to the yaw velocity value sequence of gyroscope output, according to the yaw velocity value sequence after compensation
Integral result judge whether lane-changing intention.
3. main and side road recognition methods as claimed in claim 2, which is characterized in that the foundation is to the yaw velocity after compensation
The result of value sequence integration judges whether lane-changing intention, is specially：
Mean filter processing is carried out to the yaw velocity value sequence after integration, is accumulated according to filtered yaw velocity value sequence
Point result judges whether lane-changing intention.
4. main and side road recognition methods as described in claim 1, which is characterized in that described according to the membership function, difference
The degree of membership of the path adaptation similarity maximum difference and course angle accumulated change value is calculated, is specially：
Construct the set of modes of straight-going mode and lane change pattern composition；
Using the course angle accumulated change value and path adaptation similarity maximum difference as input construction feature vector, calculate each defeated
Enter the degree of membership of construction feature vector.
5. main and side road recognition methods as claimed in claim 4, which is characterized in that described according to the degree of membership, calculating includes
The Comprehensis pertaining of straight-going mode and lane change pattern, specially：
It is worth corresponding weighted value, meter according to the default path adaptation similarity maximum difference and course angle accumulated change
Calculation includes straight-going mode and the Comprehensis pertaining of lane change pattern；
The Comprehensis pertaining is analyzed according to maximum subjection principle, determines that vehicle is in straight-going mode also in window phase
It is lane change pattern.
6. main and side road identifying system, which is characterized in that including：
Presetting module, for according to historical empirical data, presetting course angle accumulated change value and path adaptation similarity sequence most
The membership function of big difference；
Matching module, for by vehicle match to map datum road network；
First acquisition module enters window phase for working as vehicle, and the window phase is with the nearest main and side road bifurcated of vehicle front
The scope drawn a circle to approve centered on point；According to the current GPS point obtained in real time and the path adaptation similarity of matched road, obtain
The path adaptation similarity maximum difference of adjacent GPS point in window phase；
Second acquisition module, for the acquisition course angle in the yaw velocity value change sequence window phase according to gyroscope acquisition
Accumulated change value；
First computing module, for according to the membership function, calculate respectively the path adaptation similarity maximum difference and
The degree of membership of course angle accumulated change value；
Second computing module, for according to the degree of membership, calculating to include straight-going mode and the Comprehensis pertaining of lane change pattern.
7. main and side road identifying system as claimed in claim 6, which is characterized in that second acquisition module includes：
Bend compensating unit, the yaw velocity value sequence for being exported to gyroscope carry out bend compensation；
Judging unit judges whether the integral result of the yaw velocity value sequence after compensation lane-change is anticipated for foundation
Figure.
8. main and side road identifying system as claimed in claim 7, which is characterized in that it is single that second acquisition module further includes filtering
Member, for carrying out mean filter processing to the yaw velocity value sequence after integration；
The judging unit, specifically for judging whether lane-change according to filtered yaw velocity value sequence integral result
It is intended to.
9. main and side road identifying system as claimed in claim 6, which is characterized in that first computing module includes：
Structural unit, for constructing the set of modes of execution pattern and lane change pattern composition；
Computing unit, for using the course angle accumulated change value and path adaptation similarity maximum difference as input construction feature
Vector calculates the degree of membership of each input construction feature vector.
10. main and side road identifying system as claimed in claim 9, which is characterized in that second computing module includes：
Default unit, for each right according to the default path adaptation similarity maximum difference and course angle accumulated change value
The weighted value answered, calculating include straight-going mode and the Comprehensis pertaining of lane change pattern；
Analytic unit for being analyzed according to maximum subjection principle the Comprehensis pertaining, determines vehicle in window phase
In straight-going mode or lane change pattern.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611058670.4A CN108106619A (en) | 2016-11-25 | 2016-11-25 | Main and side road recognition methods and its system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611058670.4A CN108106619A (en) | 2016-11-25 | 2016-11-25 | Main and side road recognition methods and its system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108106619A true CN108106619A (en) | 2018-06-01 |
Family
ID=62204437
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201611058670.4A Pending CN108106619A (en) | 2016-11-25 | 2016-11-25 | Main and side road recognition methods and its system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108106619A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109631925A (en) * | 2018-12-21 | 2019-04-16 | 斑马网络技术有限公司 | Main and side road determines method, apparatus, storage medium and electronic equipment |
| CN115257771A (en) * | 2022-09-28 | 2022-11-01 | 天津所托瑞安汽车科技有限公司 | Intersection identification method, electronic device and storage medium |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20080021638A1 (en) * | 2006-07-18 | 2008-01-24 | Denso Corporation | Navigation system |
| CN102175254A (en) * | 2011-01-13 | 2011-09-07 | 北京超图软件股份有限公司 | Navigation positioning correction method and device, and positioning navigation system |
| CN103026176A (en) * | 2010-07-22 | 2013-04-03 | 高通股份有限公司 | Apparatus and methods for calibrating dynamic parameters of a vehicle navigation system |
| CN104101878A (en) * | 2014-06-23 | 2014-10-15 | 北京智华驭新汽车电子技术开发有限公司 | Vehicle front target recognition system and recognition method |
| CN104833361A (en) * | 2015-05-04 | 2015-08-12 | 南京邮电大学 | Multiple weight values-based map matching method under complex road conditions |
| CN105547304A (en) * | 2015-12-07 | 2016-05-04 | 北京百度网讯科技有限公司 | Road recognition method and device |
-
2016
- 2016-11-25 CN CN201611058670.4A patent/CN108106619A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20080021638A1 (en) * | 2006-07-18 | 2008-01-24 | Denso Corporation | Navigation system |
| CN103026176A (en) * | 2010-07-22 | 2013-04-03 | 高通股份有限公司 | Apparatus and methods for calibrating dynamic parameters of a vehicle navigation system |
| CN102175254A (en) * | 2011-01-13 | 2011-09-07 | 北京超图软件股份有限公司 | Navigation positioning correction method and device, and positioning navigation system |
| CN104101878A (en) * | 2014-06-23 | 2014-10-15 | 北京智华驭新汽车电子技术开发有限公司 | Vehicle front target recognition system and recognition method |
| CN104833361A (en) * | 2015-05-04 | 2015-08-12 | 南京邮电大学 | Multiple weight values-based map matching method under complex road conditions |
| CN105547304A (en) * | 2015-12-07 | 2016-05-04 | 北京百度网讯科技有限公司 | Road recognition method and device |
Non-Patent Citations (1)
| Title |
|---|
| 张涛 等: "地图匹配中快速路出入口车辆行为模式识别", 《中国公路学报》 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109631925A (en) * | 2018-12-21 | 2019-04-16 | 斑马网络技术有限公司 | Main and side road determines method, apparatus, storage medium and electronic equipment |
| CN109631925B (en) * | 2018-12-21 | 2020-11-10 | 斑马网络技术有限公司 | Main and auxiliary road determining method and device, storage medium and electronic equipment |
| CN115257771A (en) * | 2022-09-28 | 2022-11-01 | 天津所托瑞安汽车科技有限公司 | Intersection identification method, electronic device and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107492251B (en) | Driver identity recognition and driving state monitoring method based on machine learning and deep learning | |
| CN109059944B (en) | Motion planning method based on driving habit learning | |
| CN102208013B (en) | Landscape coupling reference data generation system and position measuring system | |
| CN102208012B (en) | Landscape coupling reference data generation system and position measuring system | |
| US8791996B2 (en) | Image processing system and position measurement system | |
| JP5057184B2 (en) | Image processing system and vehicle control system | |
| CN110174113A (en) | A kind of localization method, device and the terminal in vehicle driving lane | |
| CN105718750B (en) | A kind of prediction technique and system of vehicle driving trace | |
| CN102208036B (en) | Vehicle position detection system | |
| KR102613839B1 (en) | Detection of emergency vehicles | |
| CN108445503A (en) | The unmanned path planning algorithm merged with high-precision map based on laser radar | |
| CN106184220B (en) | Abnormal driving detection method in a kind of track based on vehicle location track | |
| CN110400490A (en) | Trajectory predictions method and apparatus | |
| CN113155145B (en) | Lane-level path planning method for automatic driving lane-level navigation | |
| CN107066953A (en) | It is a kind of towards the vehicle cab recognition of monitor video, tracking and antidote and device | |
| CN106092121A (en) | Automobile navigation method and device | |
| JP5333862B2 (en) | Vehicle position detection system using landscape image recognition | |
| JP2008250687A (en) | Feature information collection device and feature information collection method | |
| JP2006208223A (en) | Vehicle position recognition device and vehicle position recognition method | |
| CN109325390B (en) | Positioning method and system based on combination of map and multi-sensor detection | |
| CN107830865A (en) | A kind of vehicle target sorting technique, device, system and computer program product | |
| CN107389084A (en) | Planning driving path planing method and storage medium | |
| JP4775658B2 (en) | Feature recognition device, vehicle position recognition device, navigation device, feature recognition method | |
| CN112082567A (en) | Map path planning method based on combination of improved Astar and Grey wolf algorithm | |
| CN115523934A (en) | Vehicle track prediction method and system based on deep learning |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180601 |
|
| RJ01 | Rejection of invention patent application after publication |