CN109087534A - A kind of traffic conflict detection method based on vehicle driving trace - Google Patents
A kind of traffic conflict detection method based on vehicle driving trace Download PDFInfo
- Publication number
- CN109087534A CN109087534A CN201811174002.7A CN201811174002A CN109087534A CN 109087534 A CN109087534 A CN 109087534A CN 201811174002 A CN201811174002 A CN 201811174002A CN 109087534 A CN109087534 A CN 109087534A
- Authority
- CN
- China
- Prior art keywords
- conflict
- traffic
- classification
- data
- intersection
- 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
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/16—Anti-collision systems
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
- G08G1/0125—Traffic data processing
- G08G1/0133—Traffic data processing for classifying traffic situation
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/017—Detecting movement of traffic to be counted or controlled identifying vehicles
- G08G1/0175—Detecting movement of traffic to be counted or controlled identifying vehicles by photographing vehicles, e.g. when violating traffic rules
Abstract
A kind of traffic conflict detection method based on vehicle driving trace comprising the steps of: the first step classifies to traffic conflict；Based on the similitude of existing traffic conflict classification, the merging of similarity has been carried out to original classification, has kept its classification easy rationally.Second step is investigated by the traffic conflict to studied intersection and is acquired data, includes conflict distance, conflict time, conflict speed data, and arrangement data are done substance and researched and analysed.Third step classifies to existing traffic conflict and carries out initial clustering using SPSS statistical analysis software, finally carries out the conflict of four road level-crossing of road to be divided into three classes by arranging: conflict, exit ramp conflict in entrance driveway conflict, intersection；SPSS software is recycled to the data of independent three classes conflict, includes conflict time, conflict speed, conflict distance, does correlation analysis two-by-two.
Description
Technical field
The present invention relates to field of intelligent transportation technology, more specifically, in particular to a kind of based on vehicle driving trace
Traffic conflict detection method.
Background technique
With the construction of urbanization, the demand that motorized traffic is gone on a journey in cities and towns increases, the automobile traffic travelled in road
In each intersection, in the traffic order under signal control with non-signal control, part motor vehicle is former because driving
Vehicular traffic caused by more than because of situations such as, traffic offence bumps against, rear-end collision occurs.Meanwhile intersection is as road
Important component is easier to be influenced by wagon flow, the stream of people, traffic environment compared with other road places, and intersection is
The important pivot point of the road network of communication lines and the Multiple trauma of traffic accident, the peace of the safe condition of intersection to entire road traffic
Row for the national games has highly important influence.
According to statistics, in the traffic accident that road occurs, occupy very big ratio in the traffic accident that intersection occurs.
U.S. intersection and its neighbouring traffic accident account for about the 40.1% of total traffic accident；Traffic thing in intersection occurs for Japan
Therefore account for the 33.3% of total traffic accident；Britain about occupies 33%；France accounts for about 24%；Urban Transport intersection occurs
Accident accounts about the 30% of total traffic accident.It in Evaluation of Traffic Safety field is both at home and abroad for a long time using accident mostly
The method of statistics is not yet established meet the accident statistics that microcosmic step analysis needs at present, using casualty data statistics
The traffic safety performance of method analysis road has some limitations.Therefore, how to overcome data acquisition time too long, together
When collected data again can and casualty data there are certain correlation and replaceability, reach characterization road well and hand over
Logical authentic security is horizontal, is always the hot spot of traffic engineering technical field research, the traffic conflict technique (traffic
Conflict technique, TCT) it is exactly to grow up in this context.But due to the occurrence features of accident and
Accident statistics method and managerial segmental defect, affect the confidence level of traffic safety.Then the traffic conflict technique is made
It is widely used in studying as the evaluation method of non-accident statistics for a kind of feasible and applicable method.
The U.S. is since the 1950s, and take the lead in the application study for having carried out the traffic conflict technique.It is Canadian
CharlesV.Zegeer etc. evaluates having for signalized intersections green light signals delayed time system in application the traffic conflict technique in 1976
Effect property；The accident and conflict to the city big pier Sa Si regional 46 signalized intersections and non-signal intersection such as U.S. W.D.Glouz
It is investigated, accident and conflict is divided into 12 seed types, and establish model；BrianL.Allen etc. is by intersection
The analysis of conflict and impact generation process, is modified supplement to the traffic conflict technique, show that the traffic conflict technique can
The conclusion of traffic accident is predicted and evaluated by ground；EzraHauer etc. discussed the validity of the traffic conflict technique in 1984,
Analyze the difference of point estimations Sum Maximum Likelihood Estimate method, it is believed that the effect of Maximum Likelihood Estimation Method is preferable, with actual value compared with
It is close.
The traffic conflict technique in China starts from the late 1980s, at present still in its infancy, current research
Emphasis is concentrated mainly on the side such as validity, technique for investigation, the discrimination standard of Serious conflicts and grade scale of the traffic conflict technique
Face, systematicness are poor.Hunan University in 1988 is by International Academic Exchange, formal invitation chairman TCTCT doctor C.Hyden
It comes to China to give lectures, TCT is formally made referrals into China's traffic engineering circle for the first time；Publishing house of Southwest Jiaotong University publishes within 1994
By a monograph " the traffic conflict technique " for Soviet Union's translation, the traffic conflict technique has been systematically discussed；1997, Beijing University of Technology built
Traffic conflicts at plane intersection probability Distribution Model and security evaluation criteria are found；City is handed over by Southwest Jiaotong University within 1998
Logical safety is classified, and especially safe, safety, gras generally recognized as safe and dangerous four grades are divided into, and grade scale is equal when using
Ratio with the equivalent volume of traffic is mixed conflict to determine；2000, Xian Road Communication Univ. Zhou Wei, Luo Shigui delivered " road
Section traffic accident Multiple trauma conflict determination method " one text, on the basis of summarizing existing method, by analysis compare, foundation
A kind of Traffic Accidents on Road Block Spots determination method based on the traffic conflict technique, and it is applied to Zhengzhou Yellow River highway bridge
Conflict observe data, demonstrate the validity of the model；Luo Shi in 2001 is expensive, Zhou Wei has delivered " Road Traffic Conflict Technique
Research ", using the traffic conflict technique, road traffic conflict technique is defined and is classified, has studied sentencing for Serious conflicts emphatically
Other standard and grade scale have further been improved existing the traffic conflict technique theory, have been mentioned for road section traffic volume safety evaluation
A kind of new research method is supplied；2003, the traffic conflict technique and planed signal were controlled intersection by Wang Haixing, Xiao Guiping
The entrance volume of traffic combine, introduce the general of TC/MPCU (the traffic conflict quantity/crossing inlet mixing equivalent volume of traffic)
It reads, by studying the relationship of the magnitude of traffic flow and traffic conflict, attempts to utilize model, traffic conflict is extrapolated by traffic analysis
Then quantity carries out impact analysis of the magnitude of traffic flow to safety evaluation.
Most of research domestic at present also focuses on whether verifying traffic conflict is reliable, determines which kind of conflict type and thing
Therefore number there is strong correlation and establish conflict on the scaling module of accident, about traffic conflict detection research compared with
Few, Tan Tieniu etc. proposes a kind of traffic accident probability forecasting method based on threedimensional model vehicle tracking, but based on camera shooting
The traffic conflict detection of head still belongs to rare at present at home.Current some patents apply camera to vehicle be scanned from
And its kinematic parameter is obtained, and if number of patent application is 200810156777, the entitled measurement highway based on line array CCD
The method of upper vehicle Position And Velocity can detect multiple lanes with optical photographing means simultaneously；Number of patent application is
200910209481, a kind of entitled method and system that Target space position is judged by linkage of multi-cameras can accurately be determined
Position goes out 3 D stereo position of the monitoring objective in monitoring environment, to realize accurate monitoring.Number of patent application is
200910119341, the entitled vehicle speed measurement device based on linear array CCD camera can high-reliability and pinpoint accuracy
Ground measures the speed of vehicle, calculates vehicle commander by the imaging and speed of linear array CCD camera, to determine vehicle, presses phase to realize
The standard of the limited speed answered punishes over-speed vehicles respectively and hypervelocity punishment in real time provides precondition.Above-mentioned technology is just with road
Video carries out the detection of car speed and position, is not related to carrying out the video of intersection fusion to detect traffic
The method of conflict.
Summary of the invention
(1) technical problem
In order to improve traffic safety, transport capacity is improved, reduces traffic accident, in the prior art
The method that logical conflict is not built using video technique detection to intersection, therefore need objectively to the intersection of road
Traffic safety is evaluated, and using suitable comprehensive traffic method for collision management, improves the runnability and safe energy of intersection
Power.
(2) technical solution
In view of the above-mentioned problems, the contents of the present invention are to provide a kind of traffic conflict detection side based on vehicle driving trace
Method.
A kind of traffic conflict detection method based on vehicle driving trace comprising the steps of:
The first step classifies to traffic conflict.Based on the similitude of existing traffic conflict classification, to original classification
The merging of similarity has been carried out, has kept its classification easy rationally.
Second step is acquired data by traffic conflict to studied intersection investigation, include conflict away from
From, conflict time, conflict speed data, arrange data and do substance and research and analyse.
Third step, using SPSS statistical analysis software, with the distance that conflicts, the angle that conflicts is clustered as clustering target
Then analysis classifies to existing traffic conflict and carries out initial clustering, by arranging finally by four road level-crossing of road
Conflict carries out being divided into three classes: conflict, exit ramp conflict in entrance driveway conflict, intersection.Recycle SPSS software to independent
The data of three classes conflict include conflict time, conflict speed, conflict distance, do correlation analysis two-by-two, and data result is aobvious
Show the correlation of conflict speed with the time that conflicts.
It is in the prior art that driver encounters emergency situations when road traffic travels to the classification method of traffic conflict
Or the mode of braking taken when other reason, the hedging behavior carried out with this embody people in Conflict Studies
Important sexual factor, however the age of human factor and driver, gender, personality, physical reason etc. situation are all relevant.
Therefore the classification ambiguity of this method is larger, and enchancement factor is excessive, cannot be applied to well in traffic conflict research.
In view of this, in the first step, the vehicle flowed to each entrance driveway three carries out permutation and combination.
Table 1 is that road grade crossing turns to-conflict type relation table.
1 road grade crossing of table turns to-conflicts type relation table
Indicate that southern entrance driveway conflicts with what East, West, South, North entrance driveway difference turned to matrix R1, R2, R3, R4 respectively
Type relationship.
Wherein, R1=R2 T, R1=R4 T, this has all corresponded to the symmetry of four road level-crossings.
In second step, based on video recording camera method, the method supplemented by artificial factual survey effectively obtains sample, reads
Take colliding data and conflict time.The sufficiently large intersection of the volume of traffic is chosen, to guarantee the adequacy of data acquisition, video camera
It is placed on the residential building near two intersections, position is interference-free, and visual angle is good, and data acquisition is convenient.
High-order shooting, the view of obtained different angle are carried out using camera method.What the acquisition of conflict distance was taken is road
The object of reference in face takes the lay-by of intersection as reference, and zone distance is utilized in parking and takes to certain distance outside lane
For the distance of unit, interval time is obtained with manual time-keeping, zone distance obtains the speed that conflicts with the ratio of interval time.
The acquisition of conflict time is that a point more people read on video software, carries out artificial subjectivity using the time display of software and sentences
The beginning of disconnected conflict reads the time of this process to terminating.The acquisition of conflict speed is using stopwatch from being carried out on video
Judgement.
In the investigation of real road traffic conflicts at plane intersection, it is necessary to assure the sample size quantity of investigation collects investigation
Quantity must reach a certain amount just and can guarantee that the confidence level and precision of investigation result liquidate if sample size does not reach requirement
The result that several analyses is then likely to occur of dashing forward is insincere.Theoretical according to the ASSOCIATE STATISTICS of probability theory, Serious conflicts minimum is seen
Surveying sample size can be determined by formula (1).
Wherein: N is the smallest sample amount of required observation；P is the vehicle for designing certain driving direction of certain particular conflict
Account for the ratio of observed traffic flow；Q=1-p；K corresponds to the constant of confidence degree, as confidence level be 95% when, k=
1.96；E is the allowable error of traffic conflict Ratio Estimation value.
If necessary to analyze different types of survey data, then the number of collisions of each type will reach most
Small sample amount.The requirement that sample size should be met as far as possible in the process of actual traffic conflict investigation, it is full for being difficult to really
The traffic conflict classification that sufficient sample size needs should individually consider, independent analysis.
In the third step, data statistic analysis is carried out using statistic software SPSS, by researching and analysing survey data, benefit
It is carried out with SPSS software to independent three groups of colliding datas, the time that conflicts, conflict distance, does correlation point at conflict speed two-by-two
Analysis, finally carries out the conflict of four road level-crossing of road to be divided into three classes: conflicting in entrance driveway conflict, intersection, exports
Road conflict, and correlation analysis, data result display conflict speed and the time that conflicts are done two-by-two to the data of these three types conflict
Correlation.
(3) beneficial effect
The traffic conflict seriousness of science of the invention be divided into entrance driveway conflict, conflict, exit ramp conflict in intersection
Three types, the theoretical research for science provide very high confidence level.It is of the invention studies have shown that Serious conflicts with
It is had good correlation between accident, can be good at characterizing road traffic safety situation.The present invention selects the conflict time most
The determination of traffic conflict seriousness discriminant critical value is carried out with this for standard, with and Serious conflicts have more further been determined
Discrimination standard.A kind of traffic conflict detection method based on vehicle driving trace has critically important meaning to Evaluation of Traffic Safety
Justice only has deep thorough research that preferably the traffic conflict technique could be applied in safety evaluation conflicts technique
It goes.
Detailed description of the invention
Fig. 1 is step schematic diagram of the invention, and Fig. 2 is straight trip and train flow diagram in the same direction, and Fig. 3 is to turn left and train flow diagram in the same direction,
Fig. 4 is to turn right and train flow diagram in the same direction, and Fig. 5 is variation lane figure in the same direction, and Fig. 6 is straight trip and left side straight trip figure, Fig. 7 be straight trip with
Turn left figure on right side, and Fig. 8 is to turn left and right side left-hand rotation figure, and Fig. 9 is straight trip and right side right-hand rotation figure, and Figure 10 is to turn left to turn left with right side
Figure, Figure 11 are turned right with to figure of turning left, and Figure 12 is kept straight on and to figure of turning left.
Specific embodiment
Present invention will be further explained below with reference to the attached drawings and examples.
In the description of the present invention, the orientation of instructions such as term " on ", "lower", "front", "rear", " top ", " bottom " or
Positional relationship is to be based on the orientation or positional relationship shown in the drawings, and is merely for convenience of description of the present invention and simplification of the description, without
It is that the device of indication or suggestion meaning or element must have a particular orientation, be constructed and operated in a specific orientation, therefore not
It can be interpreted as limitation of the present invention.
A kind of traffic conflict detection method based on vehicle driving trace comprising the steps of: as shown in Figure 1,
The first step classifies to traffic conflict.Based on the similitude of existing traffic conflict classification, to original classification
The merging for having carried out similarity, which is closed, keeps its classification easy rationally.
The vehicle flowed to each entrance driveway three carries out permutation and combination, on the south for import, be divided into 11 kinds of conflicts, divide
Not are as follows: it is straight that straight trip changes lane, straight trip and left side with wagon flow in the same direction, left-hand rotation with wagon flow in the same direction, right-hand rotation with wagon flow in the same direction, in the same direction
It goes, keep straight on to turn left with right side, left-hand rotation and right side left-hand rotation, straight trip turn left with right side right-hand rotation, left-hand rotation and right side, turn right and to the left
Turn, keep straight on to turning left.Fig. 2 is straight trip and wagon flow in the same direction, and Fig. 3 is to turn left and wagon flow in the same direction, and Fig. 4 is to turn right and vehicle in the same direction
Stream, Fig. 5 are variation lanes in the same direction, and Fig. 6 is that straight trip is kept straight on left side, and Fig. 7 is that straight trip is turned left with right side, and Fig. 8 is left-hand rotation and right side
Turn left, Fig. 9 is that straight trip is turned right with right side, and Figure 10 is to turn left to turn left with right side, and Figure 11 is turned right with to turning left, and Figure 12 is straight
Row with to turning left.
According to making full use of existing classification, avoiding omitting, avoid duplicate principle, road grade crossing traffic is rushed
Prominent edit becomes 11 above-mentioned classes.
Second step is acquired data, including conflict distance, punching by the traffic conflict investigation to studied intersection
The data such as prominent time, conflict speed, arrangement data analysis are done substance and are researched and analysed.
Third step, using SPSS statistical analysis software, with the distance that conflicts, the angle that conflicts is clustered as clustering target
Analysis.Although this 11 class conflict formally there is no repeat, do not omit cover reason four road level-crossing
All conflict types, but there is no the relationships inherently disclosed between different conflict types；This 11 class conflict simultaneously is for point
Appoint so excessively many and diverse for difference, statistics, analysis, therefore the conflict type of this edit is clustered again, choose conflict
The Subsidiary Index of angle and conflict distance as traffic conflict clustering, using the observation after standardization, to existing
Traffic conflict status is clustered, and finally carries out the conflict of four road level-crossing of road to be divided into three classes by arranging: into
Conflict, exit ramp conflict in mouth road conflict, intersection, and correlation analysis, data knot are done two-by-two to the data of these three types conflict
The correlation of fruit display conflict speed and the time that conflicts.
Specific embodiment:
It chooses only there are two types of the section without dividing strip is flowed to, each conflict is related to the wagon flow of both direction, shares 4 kinds of punchings
Prominent state, therefore selected p, q are respectively 0.5, the allowable error of traffic conflict ratio estimate value depends on research precision, generally
Between 0.01~0.10, the minimum conflict observed required for various specific conflicts is acquired.It is obtained by formula (1) in various feelings
Under condition, the capacity of sample cannot be less than 30.
Due to the similitude between road cross level-crossing difference entrance driveway vehicular traffic conflict, some import
The conflict phenomenon of vehicle can represent the generally conflict of intersection；Embodiment chooses the too blunt Lu Silu two in the Road Foshan City Xin Gui-
It is illustrated for the survey data of the intersection of phase signal.Table 2 lists part colliding data, wherein the car type that conflicts, 1 is
Truck, 2 be motor bus, and 3 be electric vehicle, and 4 be minibus, and 5 be buggy, and 6 be car, and 7 be bus.
2 part colliding data table of table
It according to existing classification is made full use of, avoids omitting, avoid duplicate principle, by road grade crossing traffic conflict
Edit become 11 classes, although this 11 class conflict formally there is no repeat, do not omit cover four tunnel of road
All conflict types of level-crossing, but there is no the relationships inherently disclosed between different conflict types therefore to this
The conflict type of edit is clustered again, chooses conflict angle and conflict distance as traffic conflict clustering
Subsidiary Index clusters existing traffic conflict status using the observation after standardization.Utilize SPSS statistical
The K-meansCluster for analysing software executes quick clustering order, and it is whole to carry out classification preliminary classification to initial 11 class traffic conflict
11 class conflicts after reason are merged into 4 classes, have achieved the purpose that simplify, the acquisition of data is facilitated to investigate.Table 3 is initial clustering
Center, table 4 are iteration historical record, and table 5 is final cluster centre, and table 6 is clustering output as a result, table 7 is each cluster
In case number of cases, table 8 be class traffic conflict cluster result.
3 initial cluster center of table
4 iteration historical record of table
The final cluster centre of table 5
6 clustering of table exports result
| Serial number | Conflict classification | Class | Spacing |
| 1 | 1 | 1 | 5.750 |
| 2 | 2 | 1 | 4.692 |
| 3 | 3 | 1 | 5.210 |
| 4 | 5 | 4 | 4.723 |
| 5 | 6 | 2 | 5.952 |
| 6 | 7 | 3 | 4.971 |
| 7 | 8 | 3 | 4.118 |
| 8 | 9 | 4 | 4.167 |
| 9 | 10 | 3 | 3.955 |
| 10 | 11 | 2 | 3.826 |
Case number of cases in each cluster of table 7
8 class traffic conflict cluster result of table
| The first kind | 1、2、3、4 |
| Second class | 6、11 |
| Third class | 7、8、10 |
| 4th class | 5、9 |
The above cluster result can be seen that 11 class conflicts after preliminary classification arranges and be merged into 4 classes.In order to reach letter
The purpose of change facilitates the acquisition of data to investigate, and the 4th class traffic conflict that at this time should also be too small to sample size is analyzed.
Result above is compareed, by conflict angle that decussation mouth conflicts, the harmful consequences that may cause, conflict generation
Position, the averagely following summary made of conflict time, be shown in Table the comparison of 9 different classes of indexs of correlation.
The comparison of the different classes of index of correlation of table 9
| Classification | The first kind | Second class | Third class | 4th class |
| Conflict angle | 45 ° of < | 90 ° of > | 45 ° of > | 90 ° or so |
| Conflict consequence | It knocks into the back, side crash | Collision in the middle part of headstock, vehicle body | It knocks into the back, side crash | Collision in the middle part of headstock, vehicle body |
| Generate position | Same entrance driveway | In intersection | Same exit ramp | In intersection |
| Averagely conflict the time | 1.298375 | 1.22843 | 1.211394 | 1.218047 |
It makes discovery from observation, the 4th class conflict, that is, changing Lane conflict conflict angle in preliminary classification is 0-45
Degree, conflict may cause the type of accident to knock into the back, based on side crash.It is learnt in cluster result, the second class and the punching of the 4th class
Prominent to have generation inside intersection, conflict angle is larger, and the risk for the consequence that conflicts is larger, the seriousness consequence of generation
Larger, the averagely conflict time is less with other two classes, therefore, the acquisition precision to data do not do high request or observation
When the restriction of condition, the conflict of the second class and the 4th class is merged, such as 10 institute of table of the final cluster result after merging
Show, final cluster result analysis is shown in Table 11.
10 class traffic conflict cluster result of table
| The first kind | 1、2、3、4 |
| Second class | 5、6、9、11 |
| Third class | 7、8、10 |
The analysis of the final cluster result of table 11
| Classification | The first kind | Second class | Third class |
| Averagely conflict the time | 1.298375 | 1.2232385 | 1.211394 |
| Class name | With import to conflict | Conflict in intersection | It is same to export to conflict |
As can be seen from the table, be initially divided into 11 classes must conflict merged several times after be classified as 3 classes, respectively into
Conflict, exit ramp conflict in mouth road conflict, intersection.This is also corresponding with the physical division of intersection, does not have during classification
Have and be limited to some unitary variant, excessively detailed description is not carried out to some conflict, this method is relatively simple yet
Observation, do not need to carry out detailed division and classification to the conflict inside intersection, furthermore in the method to import
Road, exit ramp, intersection intramural conflict data acquisition more system.
Data statistics is carried out with SPSS, to conflict, exit ramp conflict three classes number of collisions in entrance driveway conflict, intersection
Conflict speed, conflict time, conflict distance in carry out correlation analysis, the i.e. correlation analysis-Bivariate of two variables
Process.Statistical result obtains 12~table of table 17.
Table 12 is based on conflict speed and conflict time descriptive statistics amount
| Mean value | Standard deviation | N | |
| Conflict speed | 8.44838798 | 3.851074972 | 303 |
| Conflict the time | 1.23267 | 0.149288 | 303 |
13 correlation analysis result of table
| Conflict speed | Conflict the time | ||
| Conflict speed | Pearson correlation | 1 | 0.135 |
| Conspicuousness (bilateral) | 0.019 | ||
| N | 303 | 303 | |
| Conflict the time | Pearson correlation | 0.135 | 1 |
| Conspicuousness (bilateral) | 0.019 | ||
| N | 303 | 303 |
Table 14 is based on conflict speed and conflict apart from descriptive statistic amount
| Mean value | Standard deviation | N | |
| Conflict speed | 8.44838798 | 3.851074972 | 303 |
| Conflict distance | 4.54 | 2.560 | 303 |
Table 15 is based on conflict speed and conflict distance correlation analyzes result
Table 16 is based on conflict distance and conflict time descriptive statistic amount
| Mean value | Standard deviation | N | |
| Conflict distance | 4.54 | 2.560 | 303 |
| Conflict the time | 1.23267 | 0.149288 | 303 |
Table 17 is based on conflict distance and analyzes result with conflict temporal correlation
| Conflict speed | Conflict distance | ||
| Conflict distance | Pearson correlation | 1 | -0.025 |
| Conspicuousness (bilateral) | 0.663 | ||
| N | 303 | 303 | |
| Conflict the time | Pearson correlation | -0.025 | 1 |
| Conspicuousness (bilateral) | 0.663 | ||
| N | 303 | 303 |
From correlation analysis is carried out between above three variable mutually two-by-two, conflict speed-conflict time, conflict speed-punching
Prominent distance, conflict distance-conflict time, the correlation analysis about Bivariate process of progress obtains as a result, from above-mentioned
In statistical result it can be found that conflict the correlation analysis of speed-conflict time in, shown in Pearson correlation coefficient and
Significance test result.Since its related coefficient is 0.135, the sig. of related coefficient is 0.019, is greater than 0.01 less than 0.05.
Illustrate that conflict speed and the correlation for the time that conflicts are significant, and about conflict speed and the distance that conflicts, conflict distance and punching
The related coefficient of prominent time is respectively 0.057, -0.025.Sig. coefficient is respectively 0.325 and 0.663.All have exceeded 0.01~
0.05 codomain range learns that this two groups of variables are not in correlation.Counted by SPSS, the results showed that conflict speed with
The time conflict in correlation.
By researching and analysing the survey data of the Foshan City Road the Xin Gui road-Tai Gen intersection, statisticallyd analyze using SPSS soft
Part, with the distance that conflicts, the angle that conflicts carries out clustering as clustering target, is sorted in Cluster merging original, simplifies former
Some classification, classification eases are that entrance driveway conflict, conflict in intersection, exit ramp conflict these three conflicts.Furthermore with SPSS
Software carries out clustering and uses SPSS software to independent three groups of data again, the time that conflicts, conflict speed, conflict apart from, two-by-two
Correlation analysis, the correlation of data result display conflict speed and the time that conflicts are done.
The present invention is not limited to the above-described embodiments, anyone can obtain other various shapes under the inspiration of the present invention
The product of formula.It is all according to equivalent changes and modifications within the scope of the patent application of the present invention, all should belong to and of the invention cover model
It encloses.
Claims (4)
1. a kind of traffic conflict detection method based on vehicle driving trace, it is characterised in that: comprise the steps of:
The first step classifies to traffic conflict；Based on the similitude of existing traffic conflict classification, original classification is carried out
The merging of similarity keeps its classification easy rationally；
Second step is acquired data by the traffic conflict investigation to studied intersection, includes conflict distance, punching
Prominent time, conflict speed data, arrangement data are done substance and are researched and analysed；
Third step classifies to existing traffic conflict and carries out initial clustering using SPSS statistical analysis software, final by arranging
It carries out the conflict of four road level-crossing of road to be divided into three classes: conflict, exit ramp conflict in entrance driveway conflict, intersection；
SPSS software is recycled to the data of independent three classes conflict, includes conflict time, conflict speed, conflict distance, does two-by-two
Correlation analysis.
2. a kind of traffic conflict detection method based on vehicle driving trace according to claim 1, it is characterised in that:
In the first step, based on the similitude of existing traffic conflict classification, its classification is made to the merging that original classification has carried out similarity
It is easy to be reasonable；To each entrance driveway three flow to vehicle carry out permutation and combination, be divided into 11 kinds of conflicts, be respectively as follows: straight trip with together
To wagon flow, turn left with wagon flow in the same direction, turn right with wagon flow in the same direction, to change lane, straight trip and left side straight trip, straight trip and right side in the same direction left
Turn, turn left to turn left with right side, straight trip is turned right with right side, turn left to turn left with right side, turn right with to turning left, keep straight on and to the left
Turn.
3. a kind of traffic conflict detection method based on vehicle driving trace according to claim 1, it is characterised in that:
In second step, based on video recording camera method, the method supplemented by artificial factual survey effectively obtains sample, read colliding data and
Conflict the time；The sufficiently large intersection of the volume of traffic is chosen, to guarantee the adequacy of data acquisition, video camera is placed in two intersections
On residential building near mouthful.
4. a kind of traffic conflict detection method based on vehicle driving trace according to claim 1, it is characterised in that:
In third step, with the distance that conflicts, the angle that conflicts carries out clustering as clustering target.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811174002.7A CN109087534A (en) | 2018-10-09 | 2018-10-09 | A kind of traffic conflict detection method based on vehicle driving trace |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811174002.7A CN109087534A (en) | 2018-10-09 | 2018-10-09 | A kind of traffic conflict detection method based on vehicle driving trace |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109087534A true CN109087534A (en) | 2018-12-25 |
Family
ID=64843359
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811174002.7A Pending CN109087534A (en) | 2018-10-09 | 2018-10-09 | A kind of traffic conflict detection method based on vehicle driving trace |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109087534A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110533913A (en) * | 2019-09-17 | 2019-12-03 | 辰芯科技有限公司 | Method for early warning, device, vehicle and the storage medium of vehicle collision |
| CN111488417A (en) * | 2019-01-28 | 2020-08-04 | 阿里巴巴集团控股有限公司 | Information processing method, system, device, equipment and computer storage medium |
| CN111968365A (en) * | 2020-07-24 | 2020-11-20 | 武汉理工大学 | Non-signalized intersection vehicle behavior analysis method and system and storage medium |
| CN112150046A (en) * | 2020-11-06 | 2020-12-29 | 上海交通大学 | Road intersection safety risk index calculation method |
| CN114155715A (en) * | 2022-02-07 | 2022-03-08 | 北京图盟科技有限公司 | Conflict point detection method, device, equipment and readable storage medium |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2010046063A1 (en) * | 2008-10-22 | 2010-04-29 | Continental Automotive France | Method of managing traffic information |
| CN101853575A (en) * | 2010-05-31 | 2010-10-06 | 北京航空航天大学 | Road junction traffic conflict detection and safety evaluation method based on two-dimensional laser scanners |
| WO2011044157A1 (en) * | 2009-10-06 | 2011-04-14 | Biogen Idec Ma Inc. | Heterocyclic compounds useful as pdk1 inhibitors |
| CN102254430A (en) * | 2011-06-03 | 2011-11-23 | 东南大学 | Method for distinguishing accident-prone road section by using traffic conflicts |
| CN102521438A (en) * | 2011-12-02 | 2012-06-27 | 东南大学 | Traffic conflict simulation two-stage parameter calibrating method |
| CN103093620A (en) * | 2013-01-07 | 2013-05-08 | 东南大学 | Determination method of motor vehicle traffic conflict number based on conflict traffic flow characteristics |
| CN103971519A (en) * | 2014-04-04 | 2014-08-06 | 东南大学 | System and method of using traffic conflicts for judging accident-prone sections |
| CN105243876A (en) * | 2015-11-06 | 2016-01-13 | 东南大学 | Traffic conflict severity analysis method used for interchange |
| CN107705635A (en) * | 2017-11-24 | 2018-02-16 | 吉林大学 | The traffic conflict method of discrimination of signalized intersections electric bicycle |
-
2018
- 2018-10-09 CN CN201811174002.7A patent/CN109087534A/en active Pending
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2010046063A1 (en) * | 2008-10-22 | 2010-04-29 | Continental Automotive France | Method of managing traffic information |
| WO2011044157A1 (en) * | 2009-10-06 | 2011-04-14 | Biogen Idec Ma Inc. | Heterocyclic compounds useful as pdk1 inhibitors |
| CN101853575A (en) * | 2010-05-31 | 2010-10-06 | 北京航空航天大学 | Road junction traffic conflict detection and safety evaluation method based on two-dimensional laser scanners |
| CN102254430A (en) * | 2011-06-03 | 2011-11-23 | 东南大学 | Method for distinguishing accident-prone road section by using traffic conflicts |
| CN102521438A (en) * | 2011-12-02 | 2012-06-27 | 东南大学 | Traffic conflict simulation two-stage parameter calibrating method |
| CN103093620A (en) * | 2013-01-07 | 2013-05-08 | 东南大学 | Determination method of motor vehicle traffic conflict number based on conflict traffic flow characteristics |
| CN103971519A (en) * | 2014-04-04 | 2014-08-06 | 东南大学 | System and method of using traffic conflicts for judging accident-prone sections |
| CN105243876A (en) * | 2015-11-06 | 2016-01-13 | 东南大学 | Traffic conflict severity analysis method used for interchange |
| CN107705635A (en) * | 2017-11-24 | 2018-02-16 | 吉林大学 | The traffic conflict method of discrimination of signalized intersections electric bicycle |
Non-Patent Citations (2)
| Title |
|---|
| 丛颖: "公路平面交叉口交通冲突分类及严重性划分", 《中国学位论文全文数据库》 * |
| 张渊渊，张玉，魏华波: "城市交通交叉口故障检测仿真研究", 《计算机仿真》 * |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111488417A (en) * | 2019-01-28 | 2020-08-04 | 阿里巴巴集团控股有限公司 | Information processing method, system, device, equipment and computer storage medium |
| CN111488417B (en) * | 2019-01-28 | 2023-10-24 | 阿里巴巴集团控股有限公司 | Information processing method, system, device, equipment and computer storage medium |
| CN110533913A (en) * | 2019-09-17 | 2019-12-03 | 辰芯科技有限公司 | Method for early warning, device, vehicle and the storage medium of vehicle collision |
| CN111968365A (en) * | 2020-07-24 | 2020-11-20 | 武汉理工大学 | Non-signalized intersection vehicle behavior analysis method and system and storage medium |
| CN111968365B (en) * | 2020-07-24 | 2022-02-15 | 武汉理工大学 | Non-signalized intersection vehicle behavior analysis method and system and storage medium |
| CN112150046A (en) * | 2020-11-06 | 2020-12-29 | 上海交通大学 | Road intersection safety risk index calculation method |
| CN114155715A (en) * | 2022-02-07 | 2022-03-08 | 北京图盟科技有限公司 | Conflict point detection method, device, equipment and readable storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109087534A (en) | A kind of traffic conflict detection method based on vehicle driving trace | |
| CN102136190B (en) | Dispatching management system and method for event emergency response of urban bus passenger transport | |
| CN111091720B (en) | Congestion road section identification method and device based on signaling data and floating car data | |
| CN105913661A (en) | Highway road section traffic state discrimination method based on charging data | |
| CN105788252A (en) | Urban trunk road vehicle trajectory reconstruction method based on fixed-point detector and signal timing data fusion | |
| CN104778834A (en) | Urban road traffic jam judging method based on vehicle GPS data | |
| CN108447256A (en) | Trunk road vehicle trajectory reconstruction method based on electric police and fixed point detector data fusion | |
| CN103903433A (en) | Real-time dynamic judgment method and device for road traffic state | |
| CN107705635B (en) | Method for judging traffic conflict of electric bicycles at signalized intersection | |
| CN111243338A (en) | Vehicle acceleration-based collision risk evaluation method | |
| CN107180534B (en) | The express highway section average speed estimation method of support vector regression fusion | |
| CN105513362A (en) | Method for evaluating and verifying running state of bus in area adjacent to bus stop | |
| CN105869402A (en) | Highway section speed correction method based on multiple types of floating car data | |
| CN101131796A (en) | Road traffic parameter checking device and method thereof | |
| CN111986483A (en) | Method and device for studying and judging road congestion state based on electric alarm data collision and storage medium | |
| Wang et al. | ARIMA model and few-shot learning for vehicle speed time series analysis and prediction | |
| Pi et al. | Automotive platoon energy-saving: A review | |
| Ma et al. | Conflict probability prediction and safety assessment of straight-left traffic flow at signalized intersections | |
| Wang et al. | A simulation-based comprehensive analysis for traffic efficiency and spatial distribution of risks in short weaving area of municipal interchange | |
| Zheng et al. | A novel set of driving style risk evaluation index system for UBI-based differentiated commercial vehicle insurance in China | |
| Zhang et al. | Modeling lane-change risk in urban expressway off-ramp area based on naturalistic driving data | |
| Hang et al. | Research on expressway state recognition based on Beidou positioning and navigation data | |
| BISWAS et al. | Performance Evaluation of Undivided Streets: A Clustering-based Approach | |
| Dong et al. | Risk analysis of vehicle rear-end collisions at intersections | |
| OUYANG et al. | Method to comprehensively evaluate road network traffic conditions during freeway reconstruction and extension |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20181225 |