CN108922168B - A kind of mid-scale view Frequent Accidents road sentences method for distinguishing - Google Patents
A kind of mid-scale view Frequent Accidents road sentences method for distinguishing Download PDFInfo
- Publication number
- CN108922168B CN108922168B CN201810527789.4A CN201810527789A CN108922168B CN 108922168 B CN108922168 B CN 108922168B CN 201810527789 A CN201810527789 A CN 201810527789A CN 108922168 B CN108922168 B CN 108922168B
- Authority
- CN
- China
- Prior art keywords
- road
- accident
- traffic
- unit
- safety
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Classifications
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/07—Controlling traffic signals
- G08G1/08—Controlling traffic signals according to detected number or speed of vehicles
-
- 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
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/07—Controlling traffic signals
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Traffic Control Systems (AREA)
Abstract
A kind of Frequent Accidents road method of discrimination of mid-scale view is proposed for the biggish urban road of road mileage.This method horizontally sees the road network form of unit two sides road by the way that urban road adjacent segments and intersection are combined into middle sight unit according to road Cross Section and traffic circulation feature group in the longitudinal direction in calculating.Based on middle geometry designs, road network feature, the traffic characteristic three classes data for seeing unit, consider the spatial coherence of the middle sight unit from same road, the negative Binomial Model of stochastic effects is established, and calculates its safety and space can be improved, Frequent Accidents road is differentiated.This method: 1) adjacent segments and the problem of influencing each other, overcome research dividing elements in conventional security analysis model of intersection are considered；2) influence of intersection spacing and road network form to accident is considered；3) it carries out Frequent Accidents road to combined section and intersection in mid-scale view to differentiate, more previous traditional method of discrimination has more Engineering Guidance meaning.
Description
Technical field
The present invention relates to traffic safety management field, in particular to a kind of side of mid-scale view Frequent Accidents road differentiation
Method.
Background technique
Urban road number of track-lines is more, and the magnitude of traffic flow is big, mode of transportation multiplicity, and accident number is more.In conventional security analysis, hand over
Prong is defined as intersection center and is defined as among Adjacent Intersections to region between stop line and the downstream security zone of influence, section
Part, and intersection adjacent in road network and section are often decomposed into the independent research unit of two classes and studied respectively.Mesh
Preceding Frequent Accidents road method of discrimination is based on conventional security analysis more, that is, determines single section or intersection is more as accident
Send out facility.However, urban road intersection spacing is short, this number of Shanghai is only 300m, intersection and section operation conditions phase
It mutually influences, it is difficult to separate the accident in intersection and section.And at present in practical engineering application, generally with continuous section and intersection
Mouth carries out safety improvement as a Frequent Accidents road.Thus consider adjacent segments and intersection combination carrying out safety analysis
It is particularly important.
The differentiation of Frequent Accidents road is namely based on safety and determines hazardous road.Accident-prone road section method of discrimination master at present
It is divided into three classes: accident number method, spatial analytical method and safety analysis modelling.Accident absolute number of the accident number method based on observation
Directly differentiated, including accident number method, accident rate method etc., traffic department, China is when carrying out accident-prone road section investigation work
Such method is usually used, but this method has ignored the Spatial Agglomeration and stochastic volatility of accident, easily leads to differentiation result
There is deviation.Spatial analytical method gathers feature using Spatial Data Analysis identification point, to judge the multiple road of accident, so
And this method does not consider the influence factor of accident, can not work for later period upgrading of a road and provide foundation and help.Building safety
Analysis model can be used to analyze the influence factor of accident, and carry out accident forecast, based on accident forecast value or construct other and refer to
Mark is final to differentiate Frequent Accidents road as space (Potential for Safety Improvement, PSI) can be improved safely
Road.
Summary of the invention
The purpose of the present invention is: a kind of Frequent Accidents road of mid-scale view is proposed for the biggish urban road of road mileage
Road method of discrimination.This method is by the longitudinal direction transporting urban road adjacent segments and intersection according to road Cross Section and traffic
Row feature group is combined into an entirety, which is defined as middle sight unit, and the road network of unit two sides road is horizontally seen in calculating
Form.Based on middle geometry designs, road network feature, the traffic characteristic three classes data for seeing unit, it is contemplated that in same road
The spatial coherence for seeing unit, establishes the negative Binomial Model of stochastic effects, and calculate its safety and space can be improved, to Frequent Accidents road
Road is differentiated.
The technical scheme adopted by the invention is that:
A kind of mid-scale view Frequent Accidents road sentences method for distinguishing, and steps are as follows:
Step 1: obtaining the geometry designs of all sections and intersection, traffic circulation and casualty data on road.
Highway geometrical design data are obtained using streetscape map,
Traffic flow data is obtained based on Coil Detector equipment, the speed of service in section is extracted according to floating car data,
Road traffic accident data are obtained according to " Shanghai City road traffic accident analyzing and alarming system ", and according to serious journey
Degree is classified as object damage accident, casualty accident.
Step 2: along road direction, road being divided into middle sight and studies unit.
The cross section parameter in highway geometrical design, the section speed of service, road section length obtained according to step 1, by road
K-path partition is middle sight unit, i.e., the middle cross section seen in unit, operation characteristic otherness are smaller.
Step 3: intersecting direction with road, two sides are selected 350m as the road network range for influencing road operation conditions, adopted
With the concentration of Jie's degree center metrization (being known in the art method) road network, road network is divided into grid, irregular grid, is mixed
Three kinds of road network forms of mould assembly.
Step 4: safety analysis sample data set is seen in building.
Based on the middle sight unit delimited in step 2, become based on the geometry feature of section in step 1 and intersection
Amount, traffic circulation characteristic variable and total accident number, object undermine casualty accident number and calculate each middle geometry feature change for seeing unit
Amount, traffic circulation characteristic variable and total accident number, object undermine casualty accident number, construct the sample data set of Traffic Safety Analysis.
Step 5: establishing the negative binomial safety analysis model of stochastic effects.
Step 5.1: constructing the negative binomial safety analysis model of stochastic effects for total accident number.Assuming that road section traffic volume accident number
Obey negative binomial distribution:
yij~Negbin (θij,r) (1)
The negative binomial safety analysis model equation of stochastic effects are as follows:
Wherein
yijFor the observation accident number for seeing unit in j-th of major trunk roads i, θijIt is yijDesired value, coefficient of dispersion r obey
Gamma is distributed (10-3,10-3), XijFor independent variable, β is estimation coefficient,For major trunk roads stochastic effects, Normal Distribution
(0,1/a), wherein a is precision parameter, obeys Gamma distribution (10-3,10-3).
Step 5.2: undermining casualty accident number for object and construct the negative binomial safety analysis model of double dependent variable stochastic effects.It is false
If object undermines casualty accident and obeys negative binomial distribution, yij1~Negbin (θij1,r1), yij2~Negbin (θij2,r2), it is described double
The negative binomial safety analysis model equation of dependent variable stochastic effects are as follows:
Wherein
yij1The object that unit is seen in certain damages accident number, yij2For the middle casualty accident number for seeing unit, θij1It is yij1Desired value,
andθij2It is yij2Desired value.It is stochastic effects item.uij1And uij2It is error term, Normal Distribution uij1~N (0,1/
τ), τ is accuracy parameter, τ~gamma (0.001,0.001).Xij1And Xij2It is independent variable.β and δm(m=1,2,3) for wait estimate
Count coefficient.
Step 5.3: estimating traffic safety model parameter using full bayes method.It is specific for parameter setting one first
Prior distribution, then Posterior distrbutionp is obtained in conjunction with observation data, and pass through Markov chain Monte-Carlo method (MarkovChain
Monte Carlo, MCMC) complete parameter Estimation.Theoretical frame are as follows:
Wherein, y is the accident number occurred, and θ is that accident number it is expected, L (y | θ) it is likelihood function, π (θ) is prior distribution,
π (θ | y) is the Posterior distrbutionp of θ under the conditions of given y, i.e., the accident number expectation that will occur, and ∫ L (y | θ) π (θ) d θ is observation number
According to marginal probability distribution.
Step 6: calculating safety can be improved space (Potential for Safety Improvement, PSI).Safety can
Improving space is that Bayesian Estimation accident number and similar place are averaged the desired difference of accident,
Wherein,Space can be improved for the safety of place ij accident pattern k, k=0 indicates total accident, k=1 expression thing
Damage accident, k=2 indicate casualty accident.For the accident number calculated according to formula (3) (4) according to Bayesian Estimation method,For the accident number calculated according to common Maximum Likelihood Estimation according to formula (3) (4).Indicate place ij's
Accident has exceeded similar place, and as caused by the place correlated characteristic, the thing that certain Improving Measurements can be taken to will exceed
Therefore it is reduced to average value.The calculation formula that sample data is substituted into PSI calculates all sections totality, object damage, the peace of casualty accident
Space can be improved entirely, centering is seen Frequent Accidents road and differentiated.
The invention has the advantages that
The invention proposes a kind of mid-scale view Frequent Accidents roads to sentence method for distinguishing.The advantage is that:
1) influencing each other for adjacent segments and intersection is considered, is overcome and is studied unit in conventional security analysis model and draw
The problem of dividing；
2) influence of intersection spacing and road network form to accident is considered；
3) it carries out Frequent Accidents road to combined section and intersection in mid-scale view to differentiate, more previous tradition
Method of discrimination has more Engineering Guidance meaning.
Detailed description of the invention
Fig. 1 is that unit example is seen in embodiment
Fig. 2 is embodiment anabolic process example
Fig. 3 is the middle sight unit of embodiment adjacent to road network range
Fig. 4 is that Frequent Accidents road differentiates result
Fig. 5 is flow chart of the invention
Specific embodiment
For the biggish urban road of road mileage, pass through acquisition road section and intersection geometry designs, traffic circulation
Adjacent segments and intersection according to road Cross Section feature are included that median strip, machine are non-along urban road direction by data
Dividing strip, traffic circulation feature and road range combinations are middle sight unit, are intersecting direction with road, with intersection spacing work
For road network range, the road network form of unit two sides road is seen in calculating.Based on the middle geometry designs for seeing unit, road network feature, hand over
Logical feature three classes data, it is contemplated that the spatial coherence of the middle sight unit from same road is established random for total accident number
The negative Binomial Model of effect undermines casualty accident number for object and establishes the negative Binomial Model of binary dependent variables stochastic effects, and in calculating
Space can be improved in the safety for seeing unit, differentiates to Frequent Accidents road.
The present invention is described in detail with specific embodiment below in conjunction with the accompanying drawings, and steps are as follows:
Step 1: obtaining the geometry designs of all sections and intersection, traffic circulation and casualty data on road.Utilize street
Scape map obtains highway geometrical design data, obtains traffic flow data based on Coil Detector equipment, is mentioned according to floating car data
Accident is divided into object damage accident, casualty accident according to the severity of traffic accident by the speed of service for taking section.
Step 2: along road direction, road being divided into middle sight and studies unit.It is run according to cross-sectional design parameter, road
Road is divided into middle sight and studies unit by speed, road range, i.e., in see cross section in unit, operation characteristic otherness compared with
It is small.
Step 3: intersecting direction with road, two sides are selected 350m as the road network range for influencing road operation conditions, adopted
With the concentration of Jie's degree center metrization road network, road network is divided into grid, irregular grid, three kinds of road network forms of mixed type.
Step 4: safety analysis sample data set is seen in building.It extracts each middle geometry feature variable for seeing unit, hand over
Logical operation characteristic variable and total accident number, object undermine casualty accident number, construct the sample data set of Traffic Safety Analysis.
Step 5: establishing the negative binomial safety analysis model of stochastic effects.
Step 5.1: constructing the negative binomial safety analysis model of stochastic effects for total accident number.Assuming that road section traffic volume accident number
Obey negative binomial distribution, yij~Negbin (θij, r), model equation isyijFor major trunk roads
The observation accident number of unit, θ are seen in j-th of iijIt is yijDesired value, coefficient of dispersion r obey Gamma distribution (10-3,10-3),
XijFor independent variable, β is estimation coefficient,For major trunk roads stochastic effects, (0,1/a), wherein a is precision ginseng to Normal Distribution
Number obeys Gamma distribution (10-3,10-3).
Step 5.2: undermining the casualty accident number building negative binomial traffic safety model of binary dependent variables stochastic effects for object.
Assuming that object, which undermines casualty accident, obeys negative binomial distribution,Model
Equation is Wherein yij1The object that unit is seen in certain damages accident number, yij2For the middle casualty accident number for seeing unit, θij1It is yij1's
Desired value, and θij2It is yij2Desired value.It is stochastic effects item.uij1And uij2It is error term, Normal Distribution uij1~
N (0,1/ τ), τ are accuracy parameter, τ~gamma (0.001,0.001).Xij1And Xij2It is independent variable.β and δm(m=1,2,3)
For coefficient to be estimated.
Step 5.3: estimating traffic safety model parameter using full bayes method.It is specific for parameter setting one first
Prior distribution, then Posterior distrbutionp is obtained in conjunction with observation data, and pass through Markov chain Monte-Carlo method (Markov Chain
Monte Carlo, MCMC) complete parameter Estimation.Theoretical frame isWherein y is the thing occurred
Therefore number, θ are that accident number it is expected, L (y | θ) is likelihood function, and π (θ) is prior distribution, and π (θ | y) is under the conditions of giving y after θ
Distribution is tested, i.e., the accident number that will occur expectation, ∫ L (y | θ) π (θ) d θ be the marginal probability distribution of observation data.
Step 6: calculating safety can be improved space (Potential for Safety Improvement, PSI).Safety can
Improving space is that Bayesian Estimation accident number and similar place are averaged the desired difference of accident, i.e.,
Wherein,Space can be improved for the safety of section ij accident pattern k, k=0 indicates that total accident, k=1 expression thing damage accident, k
=2 indicate casualty accident.Indicate that the accident of section ij has exceeded similar place, and by the section correlated characteristic institute
Cause, the accident that certain Improving Measurements can be taken to will exceed is reduced to average value.Sample data is substituted into PSI calculation formula,
Calculating all sections totality, object damage, the safety of casualty accident can be improved space.
Testing example is provided, further using GIS-Geographic Information System as spatial database, chooses 21, Shanghai City master
Arterial highway, including 411 sections (road between two neighboring intersection) and 411 signal-control crossings.And collect road geometry
Design, traffic circulation and traffic accident data, the test present invention.
To the detailed process of step 2 of the present invention " along road direction, road is divided into middle see and studies unit " are as follows:
Intersection and section combination step are as follows:
1) major trunk roads are interrupted when section median strip and the non-dividing strip setting of machine change.If in view of
Road is interrupted in intersection, and crossing accident is difficult to be reasonably allocated on the section of two sides, thus the starting point of assembled unit
It is road segment midpoints with terminal, as shown in Figure 1.
2) every major trunk roads section velocity contour is drawn, the major trunk roads average speed is calculated.Fig. 2 is certain major trunk roads section
Rate curve example, totally 23 sections, the average speed in each section of vertex representation, whole major trunk roads average speed are in figure
35km/h。
3) adjacent segments that will be above major trunk roads average speed are combined, and the section lower than average speed is combined one
It rises.In Fig. 2, which interrupts in section A and C, and from 0-A, speed is higher than 35km/h, and from A-C, speed is lower than 35km/h.
4) road is interrupted to reduce the speed difference in section in same assembled unit in velocity jump point.In Fig. 2, speed
Catastrophe point is B, D, E, which is divided into 6 assembled units: 0-A, A-B, B-C, C-D, D-E, E- terminal.In order to distinguish
The section Zhong Guan and microcosmic section guarantee that seeing in one includes two or more intersections in unit as far as possible.
Finally 21 major trunk roads are divided into 118 and see unit, average length 1230m.
Step 3 " intersects direction with road, two sides are selected 350m as the road network range for influencing road operation conditions, adopted
With the concentration of Jie's degree center metrization road network, road network is divided into grid, irregular grid, three kinds of road network forms of mixed type."
Detailed implementation process:
The road network morphology influence major trunk roads operation conditions of major trunk roads two sides and access, the further influence safe shape of major trunk roads
Condition.In view of first section of major trunk roads two sides and its road intersection directly affects the traffic noise prediction of major trunk roads, pass through
The average road network spacing for calculating road network in outer ring is 353m, thus 350m is respectively adopted in road network two sides range.Fig. 3 black polygon
Illustrating grey lines of each middle sight unit in road network range, black polygon is real road road network.
Based on sight sample data set X in obtained in step 4 118ijIt is special including geometry feature variable, traffic circulation
Levy variable and yijAccident number, the safety analysis model in establishment step 5 are as follows:
Model estimated result β is obtained according to bayes method Sum Maximum Likelihood Estimate method respectively, according to two methods
Estimated result calculates separatelyWithCalculating safety according to step 6 can be improved space, as follows:
FoundationSize Frequent Accidents road is ranked up, accident-prone road section is differentiated,It is larger
Road be Frequent Accidents road, differentiate result as shown in figure 4, black road is that safety can be improved the biggish road of spatial value.
Claims (1)
1. a kind of mid-scale view Frequent Accidents road sentences method for distinguishing, which is characterized in that steps are as follows:
Step 1: obtaining the geometry designs of all sections and intersection, traffic circulation and casualty data on road；Using streetscape
Figure obtains highway geometrical design data, obtains traffic flow data based on Coil Detector equipment, extracts road according to floating car data
Accident is divided into object damage accident, casualty accident according to the severity of traffic accident by the speed of service of section；
Step 2: along road direction, road being divided into middle sight and studies unit；According to cross-sectional design parameter, road operation speed
Road is divided into middle sight and studies unit by degree, road range, i.e., the middle cross section seen in unit, operation characteristic otherness are smaller；
Step 3: intersecting direction with road, two sides select 350m as the road network range for influencing road operation conditions, using Jie
Road network is divided into grid, irregular grid, three kinds of road network forms of mixed type by the concentration of degree center metrization road network；
Step 4: safety analysis sample data set is seen in building；Extract each middle geometry feature variable for seeing unit, traffic fortune
Row characteristic variable and total accident number, object undermine casualty accident number, construct the sample data set of Traffic Safety Analysis；
Step 5: establishing the negative binomial safety analysis model of stochastic effects；
Step 5.1: constructing the negative binomial traffic safety model of stochastic effects for total accident number；Assuming that road section traffic volume accident number is obeyed
Negative binomial distribution, yij~Negbin (θij, r), model equation isyiiFor major trunk roads i jth
A middle observation accident number for seeing unit, θijIt is yijDesired value, coefficient of dispersion r obey Gamma distribution (0.001,0.001), Xii
For independent variable, β is estimation coefficient,For major trunk roads stochastic effects, Normal Distribution (0,1/a), wherein a is precision parameter,
Obey Gamma distribution (0.001,0.001)；
Step 5.2: undermining the casualty accident number building negative binomial traffic safety model of binary dependent variables stochastic effects for object；Assuming that
Object undermines casualty accident and obeys negative binomial distribution, yij1~Negbin (θij1, r1), yij2~Negbin (θij2, r2), model equation
For Wherein yij1The object that unit is seen in certain damages accident number, yij2For the middle casualty accident number for seeing unit, θij1It is yij1
Desired value, θij2It is yij2Desired value；It is stochastic effects item；uij1And uij2It is error term, Normal Distribution uij1~N
(0,1/ τ), τ are accuracy parameter, τ~gamma (0.001,0.001)；Xij1And Xij2It is independent variable；β and δmFor system to be estimated
Number；
Traffic safety model parameter is estimated using full bayes method；It is first one specific prior distribution of parameter setting, then
Posterior distrbutionp is obtained in conjunction with observation data, and parameter Estimation is completed by Markov chain Monte-Carlo method；Theoretical frame isWherein y is the accident number occurred, and θ is that accident number it is expected, and L (y | θ) it is likelihood function, π (θ)
For prior distribution, the Posterior distrbutionp that π (θ | y) is θ under the conditions of given y, i.e., the accident number expectation that will occur, ∫ L (y | θ) π
(θ) d θ is the marginal probability distribution for observing data；
Step 6: calculating safety can be improved space；It is that Bayesian Estimation accident number and similar place are averaged thing that space, which can be improved, in safety
Therefore desired difference, i.e.,Wherein,For place ij accident pattern k's
Space can be improved in safety, and k=0 indicates that total accident, k=1 expression thing damage accident, and k=2 indicates casualty accident；Indicate ground
The accident of point ij has exceeded similar place, and as caused by the place correlated characteristic, the thing that Improving Measurements can be taken to will exceed
Therefore it is reduced to average value；Sample data is substituted into calculation formula, calculating all places totality, object damage, the safety of casualty accident can
Space is improved, to differentiate Frequent Accidents road.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810527789.4A CN108922168B (en) | 2018-05-29 | 2018-05-29 | A kind of mid-scale view Frequent Accidents road sentences method for distinguishing |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810527789.4A CN108922168B (en) | 2018-05-29 | 2018-05-29 | A kind of mid-scale view Frequent Accidents road sentences method for distinguishing |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108922168A CN108922168A (en) | 2018-11-30 |
| CN108922168B true CN108922168B (en) | 2019-10-18 |
Family
ID=64418245
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810527789.4A Active CN108922168B (en) | 2018-05-29 | 2018-05-29 | A kind of mid-scale view Frequent Accidents road sentences method for distinguishing |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108922168B (en) |
Families Citing this family (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110599769B (en) * | 2019-09-10 | 2022-08-09 | 南京城建隧桥经营管理有限责任公司 | Hierarchical ranking method for road importance in urban road network in time intervals |
| CN112419709B (en) * | 2020-10-16 | 2021-11-09 | 同济大学 | Expressway accident risk real-time prediction method based on road section heterogeneity |
| CN113887569B (en) * | 2021-09-09 | 2022-10-25 | 同济大学 | Accident frequent region distinguishing method based on region scale |
| CN113919144A (en) * | 2021-09-27 | 2022-01-11 | 同济大学 | Mountain area highway accident cause analysis model modeling method and storage medium |
| CN113920725B (en) * | 2021-09-30 | 2022-12-16 | 同济大学 | Urban suburban main road accident prediction method based on mesoscopic level and storage medium |
| CN113920738B (en) * | 2021-10-26 | 2023-04-21 | 北京工业大学 | Urban trunk security analysis model migration method based on different country merging data |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101807224A (en) * | 2010-03-24 | 2010-08-18 | 上海交通大学 | Mesoscopic-microcosmic integrated traffic simulation vehicle flow loading method |
| CN101833610A (en) * | 2010-04-09 | 2010-09-15 | 北京工业大学 | Accident black-spot identification optimizing method |
| 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 |
| CN102103715A (en) * | 2010-11-18 | 2011-06-22 | 上海海事大学 | Negative binomial regression-based maritime traffic accident investigation analysis and prediction method |
| CN106875687A (en) * | 2017-04-24 | 2017-06-20 | 哈尔滨工业大学 | A kind of accident-prone road section automatic identification method based on sliding window method |
-
2018
- 2018-05-29 CN CN201810527789.4A patent/CN108922168B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101807224A (en) * | 2010-03-24 | 2010-08-18 | 上海交通大学 | Mesoscopic-microcosmic integrated traffic simulation vehicle flow loading method |
| CN101833610A (en) * | 2010-04-09 | 2010-09-15 | 北京工业大学 | Accident black-spot identification optimizing method |
| 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 |
| CN102103715A (en) * | 2010-11-18 | 2011-06-22 | 上海海事大学 | Negative binomial regression-based maritime traffic accident investigation analysis and prediction method |
| CN106875687A (en) * | 2017-04-24 | 2017-06-20 | 哈尔滨工业大学 | A kind of accident-prone road section automatic identification method based on sliding window method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108922168A (en) | 2018-11-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108922168B (en) | A kind of mid-scale view Frequent Accidents road sentences method for distinguishing | |
| CN109544932B (en) | Urban road network flow estimation method based on fusion of taxi GPS data and gate data | |
| CN109448369B (en) | Real-time operation risk calculation method for expressway | |
| Debnath et al. | Navigational traffic conflict technique: a proactive approach to quantitative measurement of collision risks in port waters | |
| CN102360525B (en) | Discriminant analysis-based high road real-time traffic accident risk forecasting method | |
| CN104408925B (en) | Crossing evaluation of running status method based on display radar | |
| CN104751642B (en) | A kind of advanced road real-time predictor method of traffic flow operation risk | |
| CN110164132B (en) | Method and system for detecting road traffic abnormity | |
| Wang et al. | Crash estimation at signalized intersections: significant factors and temporal effect | |
| CN102360526B (en) | Real-time monitoring method for road section state of high road | |
| CN116644373B (en) | Automobile flow data analysis management system based on artificial intelligence | |
| CN104809879A (en) | Expressway road traffic state estimation method based on dynamic Bayesian network | |
| CN105825669A (en) | System and method for identifying urban expressway traffic bottlenecks | |
| CN104732075A (en) | Real-time prediction method for urban road traffic accident risk | |
| CN114783183A (en) | Monitoring method and system based on traffic situation algorithm | |
| CN111179592B (en) | Urban traffic prediction method and system based on spatio-temporal data flow fusion analysis | |
| CN105513362B (en) | A kind of bus platform adjacent area bus running state evaluation verification method | |
| CN113033840B (en) | Method and device for judging highway maintenance | |
| CN100481153C (en) | Method for automatically inspecting highway traffic event based on offset minimum binary theory | |
| CN106526149B (en) | A kind of Pavement Condition prediction technique based on be open to traffic duration and the volume of traffic | |
| CN109598931A (en) | Group based on traffic safety risk divides and difference analysis method and system | |
| CN113935158A (en) | Traffic detector optimized layout method for multiple points of highway accidents | |
| CN104575049B (en) | A kind of elevated ramp intellectual inducing method and device based on array radar | |
| CN109710720A (en) | A kind of crime risk such environmental effects analysis method for taking background effect into account | |
| CN111915880A (en) | Method for estimating effective traffic capacity of accident point of highway abnormal event |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |