CN106652441A - Urban road traffic condition prediction method based on spatial-temporal data - Google Patents
Urban road traffic condition prediction method based on spatial-temporal data Download PDFInfo
- Publication number
- CN106652441A CN106652441A CN201510739971.2A CN201510739971A CN106652441A CN 106652441 A CN106652441 A CN 106652441A CN 201510739971 A CN201510739971 A CN 201510739971A CN 106652441 A CN106652441 A CN 106652441A
- Authority
- CN
- China
- Prior art keywords
- traffic
- time
- road
- data
- space
- 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
Abstract
The invention discloses an urban road traffic condition prediction method based on spatial-temporal data. The method comprises the following steps: calculating the parameters of a spatial-temporal correlation model using historical traffic data; abstracting an urban road network in the form of undirected graph; calculating the weight of the undirected graph using historical data; building a time domain correlation model; building a spatial-temporal correlation model; and predicting the traffic condition of a road section through use of real-time traffic data and based on a time-space domain model. A more accurate urban road traffic condition prediction method is provided.
Description
Technical field
The present invention relates to a kind of urban road traffic state Forecasting Methodology based on space-time data, belongs to intelligent transportation system research neck
Domain.
Background technology
Intelligent transportation system has been successfully applied to the multiple fields such as traffic administration, induced travel, signal control, safe driving.
Traffic behavior real-time estimation and traffic behavior real-time estimate are two kinds of technologies important in intelligent transportation system.Traffic behavior is estimated in real time
Meter is typically with being deployed in the real time data of road surface sensor transmission, estimates the traffic behavior of road.What traffic status prediction referred to
It is the traffic data using real-time estimation, and the rule obtained by analysis of history data, predict future time period traffic behavior
Technology.
The related research work of traffic status prediction includes：The history method of average is a kind of the most commonly used traffic status prediction method,
It adopts the average of historical data for the estimate of current traffic condition；Yao Zhisheng, Shao Chunfu of Beijing Jiaotong University is Chinese public
Deliver on the journal of road《Road traffic state multi-spot time series based on state-space model are predicted》Propose in paper and be based on
The road traffic state multi-spot time series Forecasting Methodology of state-space model.First, using the multiple spot time of road traffic state
Sequence data sets up multidimensional autoregressive model, and conversion conditions spatial model form obtains multiple spot road traffic followed by EM algorithms
The state-space model of state；Finally, using the multi-spot time series of adjacent 6 traffic detectors collection in certain city expressway
The validity of data verification model.Sun Zhanquan, Liu Wei, Zhu Xiaomin of Shandong Province Computing Center exists《Mass transportation stream prediction side
Method research》Propose a kind of methods of sampling based on stratified sampling in combination with k mean clusters in paper, and with based on it is sequential most
The SVMs of little optimization method is combined, and carries out mass transportation stream prediction.Beijing Jiaotong University's city complexity traffic system reason
By Zhang Xinzhe, Guan Wei with technical education portion《Traffic status prediction method based on multi-parameter state for time sequence》In paper
The method that the multidimensional space data for belonging to each state is converted into One-dimension Time Series is proposed, for this state for time sequence is adopted
The traffic status prediction of lower 1 period has been carried out with BP neural network, and algorithm is simple, with stronger prediction real-time.
Said method often only considers correlation of the traffic behavior in time-domain (same section different periods), seldom considers
Correlation on different spaces domain (same period different sections of highway).And traffic behavior is to a certain extent both by the shadow of time-domain
Ring, and affected by different spaces domain, only when the two R. concomitans, the higher traffic status prediction of the degree of accuracy could be obtained.
This patent is exactly the time that considered and spatially effect of the different information to predicting, is conducive to improving the standard of traffic status prediction
True rate.
In Publication No. CN102087787A《Short-time traffic state predicting device and Forecasting Methodology》In patent, propose by from
Historical traffic status data is partitioned into some traffic state datas, accurately characterizes the traffic behavior characteristic of each traffic state data,
And the method by being predicted using fuzzy averaging, impact of the traffic behavior non-stationary change to predicting the outcome is reduced, it is big so as to realize
The application of scale ground.In Publication No. CN103413443A《Short-term traffic flow trend prediction method based on HMM》
In patent, the data for collecting are constituted first hidden state and the observation state set of HMM, then, utilized
Viterbi algorithm tries to achieve the hidden status switch of optimum, then the last state of optimum hidden status switch is predicted state.More than
Method illustrated by two patents only considered impact of the time-domain to traffic status prediction, and this patent has considered the time
Spatially effect of the different information to predicting, more effectively improves the accuracy of traffic status prediction.
Space-time data refers to the data collected in extensive spatial dimension and time range.For traffic application, refer to
It is the data in a large amount of periods of a feature road network.
Non-directed graph is that side does not have directive figure, including point, side and the set of relation between the two, is one kind weight of data structure
Want form.
Least square method is a kind of mathematical optimization techniques, can be used for parameter Estimation and curve matching.It is by minimizing error
Quadratic sum finds the optimal function matching of data.Unknown data can be easily tried to achieve using least square method, and causes these
The quadratic sum of the error between the data tried to achieve and real data is minimum.
Traffic congestion index, also known as traffic circulation index (Traffic Performance Index (TPI)), is that reflection road network is smooth
The conceptual numerical value of logical or congestion, referred to asTraffic index, can be used to represent road traffic state.Traffic index span is 0
To 10, every grade of 2 number one corresponds to respectively " unimpeded ", " substantially unimpeded ", " slight congestion ", " moderate congestion ", " serious
Five ranks of congestion ", numerical value is higher, shows that traffic congestion is more serious.
The content of the invention
The invention provides a kind of urban road traffic state Forecasting Methodology based on space-time data, by modeling the big rule of whole road network
Traffic data in mould time range, obtains correlation of the traffic behavior in time-domain and spatial domain to carry out road traffic
Status predication, improves the accuracy of prediction.
For achieving the above object, the present invention is divided into two parts of space time correlation model parameter calculation and traffic status prediction, main skill
Art scheme includes：
Step 1：The parameter of space time correlation model is calculated using a large amount of historical traffic datas.Traffic data is passed using the traffic of deployment
Sensor is gathered, and traffic sensor can be using Floating Car GPS, microwave, bayonet socket and coil etc. in this programme.By to a large amount of history
The calculating of traffic data, sets up same section between different periods, and between same period different sections of highway on traffic behavior
Correlation.Specifically include：
Step 1.1：The abstract city road network in the form of non-directed graph.According to the syntople between section, all roads in road network
Section can be expressed as undirected diagram form G=(V, E, W), wherein, V is vertex set, and i-th section is abstracted into summit in road network
vi(vi∈V)；E is line set, side eI, jRepresent viAnd vjTwo section is joined directly together, and there is syntople；W is weight set,
Weight wI, jRepresent viTo vjTraffic behavior influence degree.The correlation between adjacent segments in road network is thus established,
Namely spatial relationship.
Step 1.2：The weight of non-directed graph is calculated using historical data.Calculating the correlation of different sections of highway synchronization is used for table
Show weight wI, j, computing formula is as follows：
Wherein, wI, jRepresent viTo vjTraffic behavior influence degree, sI, tRepresent node viIn the traffic behavior of period t, N is represented and adopted
The period total number that all historical datas of collection are included.Period refers to the time cycle of estimating road traffic state, such as 5 points
Clock, 15 minutes.
Step 1.3：Build time-domain correlation model.Consider preamble period traffic behavior to current impact.Same section is different
Period traffic behavior correlation, is expressed as follows：
Wherein, sI, tRepresent node viIn the traffic behavior of period t, P represents total when hop count,For node viThe traffic of p-th period
Impact coefficient of the state to current traffic condition.When a large amount of historical datas for collecting same section, least square can be adopted
Method is calculated
Step 1.4：Build space time correlation model.Consider section in incidence relation present on time-domain and spatial domain simultaneously.Together
The correlation of one section different time traffic behavior, is expressed as follows：
Wherein, sI, tRepresent node viIn the traffic behavior of period t, M represents the related road number of the present road chosen, wI, jRepresent
viTo vjTraffic behavior influence degree, α and β is respectively the proportion in time and space, and P represents total when hop count,For section
Point viImpact coefficient of the traffic behavior of p-th period to current traffic condition.
Step 2：Road section traffic volume status predication is carried out based on space-time domain model using real time traffic data.
Step 2.1：Space-time model parameter Estimation.Related roads are the sections for having certain correlation with current road segment, that is, wI, j
It is larger.From vertex v in figure GiSet out, v is accessed firstiEach abutment points not accessed, if wI, j＞ W, vjIt is chosen
For related roads, W is the mean value of direct neighbor road correlation；Then from these abutment points them are accessed successively respectively
The abutment points not accessed, if wI, k=wI, j*wJ, k＞ 0.5c* W, wherein c are the level for accessing, and direct neighbor is the
0 layer；Until all nodes are all accessed to.After completing related roads selection, according to formula (3), using historical data, adopt
α and β is calculated with least square method.
Step 2.2：Traffic status prediction.When related roads have completed to select in spatial domain, and determine time-domain and sky
Between domain proportion, can adopt formula (3) predicted link traffic behavior.Related roads are selected and time and space proportion is arranged all
It is to be calculated using historical data.When traffic status prediction, the road traffic state obtained using present period is needed.
Description of the drawings
Fig. 1 is the flow chart of the present invention.
Specific embodiment
In order that the objects, technical solutions and advantages of the present invention become more apparent, below with reference to drawings and Examples, to this
Invention is further described.
Referred in the traffic congestion of 2015.05.01 with the section on city of Hangzhou Bei Shanlu-BaoChu Lu-dawn road and its neighbouring road
As a example by number is as traffic behavior.
Step 101, the abstract city road network in the form of non-directed graph.According to the syntople between section, all roads in road network
Section can be expressed as undirected diagram form G=(V, E, W), wherein, V is vertex set, and i-th section is abstracted into summit in road network
vi(vi∈V)；E is line set, side eI, jRepresent viAnd vjTwo section is joined directly together, and there is syntople；W is weight set,
Weight wI, jRepresent viTo vjTraffic behavior influence degree.The correlation between adjacent segments in road network is thus established,
Namely spatial relationship.
It should be noted that section here refers to the traffic route in transportation network between two neighboring intersection.
Step 102, if current road segment is node v0, road circuit node adjacent thereto is respectively v1、v2、v3、v4、v5、v6,
They are respectively w with the correlation of current road segment node0,1、w0,2、w0,3、w0,4、w0,5、w0,6, by formula (1), obtain
To such as following table：
| w0,1 | w0,2 | w0,3 | w0,4 | w0,5 | w0,6 |
| 0.3 | 0.2 | 0.05 | 0.1 | 0.18 | 0.08 |
Step 103, the mean value W that direct neighbor road correlation can be obtained by step 102 is 0.152, if wI, j＞ W, vjQuilt
Select as related roads, otherwise be then uncorrelated road.The present embodiment pair will be carried out with the road of these related roads direct neighbors
Its correlation calculations with current road segment, by formula (1), obtains such as following table：
| w0,7 w | 0,8 | w0,9 | w0,10 | w0,11 | w0,12 | w0,13 | w0,14 | w0,15 |
| 0.16 | 0.12 | 0.03 | 0.14 | 0.05 | 0.01 | 0.1 | 0.2 | 0.11 |
Correlation to more than, carries out again wI, jWith the comparison of W, draw related roads, pair with these related roads direct neighbors
Road carry out its correlation calculations with current road segment again, by formula (1), obtain such as following table：
| w0,16 | w0,17 | w0,18 | w0,19 | w0,20 | w0,21 |
| 0.12 | 0.09 | 0.06 | 0.01 | 0.02 | 0.1 |
So far, related roads never again, this step is without the need for down carry out again.
Related roads obtained as above are integrated, is obtained such as following table：
| w0,1 | w0,2 | w0,5 | w0,7 | w0,14 |
| 0.3 | 0.2 | 0.18 | 0.16 | 0.2 |
According to the present period that road surface sensor is collected, the congestion index of related roads is as follows：
| v1 | v2 | v5 | v7 | v14 | |
| Congestion index | 4 | 6 | 3 | 5 | 4 |
Using the data of two above form, can be obtained according to formula (3)：
Step 104, structure time-domain correlation model, the traffic congestion index of the current road segment collected according to road surface sensor,
Obtain such as following table：
| sI, 1 | sI, 2 | sI, 3 | sI, 4 |
| 3 | 5 | 4 | 5 |
Influence degree such as following table of the historical traffic congestion index of its five periods of current road segment to current road segment：
Using the data of two above form, calculated according to formula (2), can be obtained：
Step 105, builds space time correlation model.Through data analysis, it is 60% that can obtain α for 40%, β, according to step 103 and
Step 104, according to formula (3), can be predicted out the traffic congestion index of current road segment, specific as follows：
In addition to above-mentioned single step is tested, performance comparison analysis is We conducted.In experiment, the same area in Hangzhou civic center area is acquired
140 day datas in 50, domain section, first calculate traffic congestion index, as right value；Then selecting wherein 10 is used for property
Can estimate.The method that we compare the history method of average and this patent.If it is identical with actual result to predict the outcome, prediction is being represented just
Really, otherwise prediction error.Experimental result is that the accuracy rate of the history method of average is 79.2%, and this patent method is 88.3%.This says
The bright present invention's is a kind of higher traffic status prediction side of degree of accuracy based on the urban road traffic state Forecasting Methodology of space-time data
Method.
Claims (4)
1. a kind of urban road traffic state Forecasting Methodology based on space-time data, it is characterised in that include the step of the method：
The parameter of space time correlation model is calculated using a large amount of historical traffic datas；
(1) the abstract city road network in the form of non-directed graph；
(2) weight of non-directed graph is calculated using historical data；
(3) time-domain correlation model is built；
(4) space time correlation model is built；
(5) road section traffic volume status predication is carried out based on space-time domain model using real time traffic data.
2. according to claim 1, the weight method of non-directed graph is calculated using historical data, it is characterised in that：Using history
Data calculate the weight of non-directed graph, and calculating the correlation of different sections of highway synchronization is used to represent weight.
3. according to claim 1, space time correlation model method is built, it is characterised in that：Consider section in time-domain simultaneously
With incidence relation present on spatial domain, the correlation of same section different time traffic behavior represents as follows：
Wherein, sI, tRepresent node viIn the traffic behavior of period t, M represents the related road number of the present road chosen, wI, jRepresent
viTo vjTraffic behavior influence degree, α and β is respectively the proportion in time and space, and P represents total when hop count,For section
Point viImpact coefficient of the traffic behavior of p-th period to current traffic condition.
4. according to claim 1, road section traffic volume trend prediction method is carried out based on space-time domain model using real time traffic data,
It is characterized in that：From vertex v in figure GiSet out, v is accessed firstiEach abutment points not accessed, if wI, j＞ W, vj
Be selected as related roads, W for direct neighbor road correlation mean value, wI, jRepresent viTo vjTraffic behavior impact journey
Degree, then from these abutment points accesses successively their abutment points not accessed respectively, if
wI, k=wI, j*wJ, k＞ 0.5c* W, wherein c are the level for accessing, and direct neighbor is the 0th layer, until all nodes are all interviewed
Ask.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510739971.2A CN106652441A (en) | 2015-11-02 | 2015-11-02 | Urban road traffic condition prediction method based on spatial-temporal data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510739971.2A CN106652441A (en) | 2015-11-02 | 2015-11-02 | Urban road traffic condition prediction method based on spatial-temporal data |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN106652441A true CN106652441A (en) | 2017-05-10 |
Family
ID=58851054
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510739971.2A Pending CN106652441A (en) | 2015-11-02 | 2015-11-02 | Urban road traffic condition prediction method based on spatial-temporal data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106652441A (en) |
Cited By (18)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107230351A (en) * | 2017-07-18 | 2017-10-03 | 福州大学 | A kind of Short-time Traffic Flow Forecasting Methods based on deep learning |
| CN107730925A (en) * | 2017-11-01 | 2018-02-23 | 浙江大学城市学院 | A kind of supersaturated traffic intersection method for optimally controlling based on state space target curve |
| CN108022425A (en) * | 2017-12-21 | 2018-05-11 | 东软集团股份有限公司 | Traffic movement prediction method, device and computer equipment |
| CN108629974A (en) * | 2018-05-17 | 2018-10-09 | 电子科技大学 | Take the traffic circulation index method for building up of urban road traffic network feature into account |
| CN108629979A (en) * | 2018-06-12 | 2018-10-09 | 浙江工业大学 | A kind of congestion prediction algorithm based on history and junction perimeter data |
| CN109064750A (en) * | 2018-09-28 | 2018-12-21 | 深圳大学 | Urban road network traffic estimation method and system |
| CN109087509A (en) * | 2018-09-04 | 2018-12-25 | 重庆交通大学 | A kind of road grid traffic operating status prediction technique |
| CN109190948A (en) * | 2018-08-20 | 2019-01-11 | 北京航空航天大学 | A kind of association analysis method of large aerospace hinge operation and urban traffic blocking |
| CN109919347A (en) * | 2017-12-13 | 2019-06-21 | 腾讯科技（深圳）有限公司 | Road conditions generation method, relevant apparatus and equipment |
| CN110415521A (en) * | 2019-07-31 | 2019-11-05 | 京东城市(北京)数字科技有限公司 | Prediction technique, device and the computer readable storage medium of traffic data |
| CN110751828A (en) * | 2019-09-10 | 2020-02-04 | 平安国际智慧城市科技股份有限公司 | Road congestion measuring method and device, computer equipment and storage medium |
| CN111428920A (en) * | 2020-03-17 | 2020-07-17 | 北京香侬慧语科技有限责任公司 | Traffic prediction method and system based on graph attention network |
| CN111710155A (en) * | 2020-05-21 | 2020-09-25 | 东莞数汇大数据有限公司 | Method and system for analyzing congestion relationship between roads |
| CN111768618A (en) * | 2020-06-04 | 2020-10-13 | 北京航空航天大学 | Traffic jam state propagation prediction and early warning system and method based on city portrait |
| CN112185108A (en) * | 2020-08-27 | 2021-01-05 | 银江股份有限公司 | Urban road network congestion mode identification method, equipment and medium based on space-time characteristics |
| CN112581752A (en) * | 2019-12-30 | 2021-03-30 | 西安金路交通工程科技发展有限责任公司 | Active intervention control method and system for regional road network traffic coordination |
| WO2021073524A1 (en) * | 2019-10-15 | 2021-04-22 | 同济大学 | Analysis method for tracing source of congestion traffic flow |
| CN114648880A (en) * | 2022-05-24 | 2022-06-21 | 阿里巴巴达摩院(杭州)科技有限公司 | Method for predicting traffic flow, vehicle and readable storage medium |
-
2015
- 2015-11-02 CN CN201510739971.2A patent/CN106652441A/en active Pending
Cited By (26)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107230351A (en) * | 2017-07-18 | 2017-10-03 | 福州大学 | A kind of Short-time Traffic Flow Forecasting Methods based on deep learning |
| CN107730925A (en) * | 2017-11-01 | 2018-02-23 | 浙江大学城市学院 | A kind of supersaturated traffic intersection method for optimally controlling based on state space target curve |
| US11423774B2 (en) | 2017-12-13 | 2022-08-23 | Tencent Technology (Shenzhen) Company Limited | Road condition generation method, apparatus and device, and storage medium |
| CN109919347A (en) * | 2017-12-13 | 2019-06-21 | 腾讯科技（深圳）有限公司 | Road conditions generation method, relevant apparatus and equipment |
| CN109919347B (en) * | 2017-12-13 | 2022-01-18 | 腾讯科技（深圳）有限公司 | Road condition generation method, related device and equipment |
| CN108022425B (en) * | 2017-12-21 | 2020-05-05 | 东软集团股份有限公司 | Traffic condition prediction method and device and computer equipment |
| CN108022425A (en) * | 2017-12-21 | 2018-05-11 | 东软集团股份有限公司 | Traffic movement prediction method, device and computer equipment |
| CN108629974A (en) * | 2018-05-17 | 2018-10-09 | 电子科技大学 | Take the traffic circulation index method for building up of urban road traffic network feature into account |
| CN108629979A (en) * | 2018-06-12 | 2018-10-09 | 浙江工业大学 | A kind of congestion prediction algorithm based on history and junction perimeter data |
| CN109190948A (en) * | 2018-08-20 | 2019-01-11 | 北京航空航天大学 | A kind of association analysis method of large aerospace hinge operation and urban traffic blocking |
| CN109190948B (en) * | 2018-08-20 | 2020-06-12 | 北京航空航天大学 | Correlation analysis method for operation of large-scale aviation hub and urban traffic jam |
| CN109087509A (en) * | 2018-09-04 | 2018-12-25 | 重庆交通大学 | A kind of road grid traffic operating status prediction technique |
| CN109064750A (en) * | 2018-09-28 | 2018-12-21 | 深圳大学 | Urban road network traffic estimation method and system |
| CN109064750B (en) * | 2018-09-28 | 2021-09-24 | 深圳大学 | Urban road network traffic estimation method and system |
| CN110415521A (en) * | 2019-07-31 | 2019-11-05 | 京东城市(北京)数字科技有限公司 | Prediction technique, device and the computer readable storage medium of traffic data |
| CN110751828A (en) * | 2019-09-10 | 2020-02-04 | 平安国际智慧城市科技股份有限公司 | Road congestion measuring method and device, computer equipment and storage medium |
| WO2021073524A1 (en) * | 2019-10-15 | 2021-04-22 | 同济大学 | Analysis method for tracing source of congestion traffic flow |
| CN112581752A (en) * | 2019-12-30 | 2021-03-30 | 西安金路交通工程科技发展有限责任公司 | Active intervention control method and system for regional road network traffic coordination |
| CN111428920A (en) * | 2020-03-17 | 2020-07-17 | 北京香侬慧语科技有限责任公司 | Traffic prediction method and system based on graph attention network |
| CN111428920B (en) * | 2020-03-17 | 2023-06-09 | 北京香侬慧语科技有限责任公司 | Traffic prediction method and system based on graph attention network |
| CN111710155A (en) * | 2020-05-21 | 2020-09-25 | 东莞数汇大数据有限公司 | Method and system for analyzing congestion relationship between roads |
| CN111768618A (en) * | 2020-06-04 | 2020-10-13 | 北京航空航天大学 | Traffic jam state propagation prediction and early warning system and method based on city portrait |
| CN111768618B (en) * | 2020-06-04 | 2021-07-20 | 北京航空航天大学 | Traffic jam state propagation prediction and early warning system and method based on city portrait |
| CN112185108B (en) * | 2020-08-27 | 2021-11-16 | 银江技术股份有限公司 | Urban road network congestion mode identification method, equipment and medium based on space-time characteristics |
| CN112185108A (en) * | 2020-08-27 | 2021-01-05 | 银江股份有限公司 | Urban road network congestion mode identification method, equipment and medium based on space-time characteristics |
| CN114648880A (en) * | 2022-05-24 | 2022-06-21 | 阿里巴巴达摩院(杭州)科技有限公司 | Method for predicting traffic flow, vehicle and readable storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106652441A (en) | Urban road traffic condition prediction method based on spatial-temporal data | |
| Kim et al. | Urban traffic flow prediction system using a multifactor pattern recognition model | |
| CN111598148B (en) | Capacity evaluation method and device based on historical capacity similarity characteristics | |
| CN106997669B (en) | Method for judging traffic jam cause based on feature importance | |
| CN108109382B (en) | Congestion point, congestion line and congestion area discovery method based on composite network | |
| CN104157139B (en) | A kind of traffic congestion Forecasting Methodology and method for visualizing | |
| CN102087788B (en) | Method for estimating traffic state parameter based on confidence of speed of float car | |
| CN102693633B (en) | Short-term traffic flow weighted combination prediction method | |
| CN106503840A (en) | Parking stall Forecasting Methodology and system can be used in parking lot | |
| CN105702029A (en) | Express way traffic state prediction method taking spatial-temporal correlation into account at different times | |
| CN110390349A (en) | Bus passenger flow volume based on XGBoost model predicts modeling method | |
| CN110836675B (en) | Decision tree-based automatic driving search decision method | |
| Zou et al. | A spatial analysis approach for describing spatial pattern of urban traffic state | |
| JP2011138487A (en) | Method and system for traffic prediction based on space-time relation | |
| CN111080029A (en) | Urban traffic road speed prediction method and system based on multi-path segment space-time correlation | |
| CN111292534A (en) | Traffic state estimation method based on clustering and deep sequence learning | |
| CN109859480B (en) | Congestion road section modeling and evaluating method based on complex network | |
| CN105405293A (en) | Short-term prediction method of road travel time and system | |
| CN104021674A (en) | Method for rapidly and accurately forecasting travel time of vehicles for passing through road sections | |
| CN109584552A (en) | A kind of public transport arrival time prediction technique based on network vector autoregression model | |
| CN109272169A (en) | Traffic flow forecasting method, device, computer equipment and storage medium | |
| CN112907970B (en) | Variable lane steering control method based on vehicle queuing length change rate | |
| CN106297296B (en) | A kind of fine granularity hourage distribution method based on sparse track point data | |
| CN106845703B (en) | Urban road network time-varying K shortest path searching method considering steering delay | |
| CN103745602B (en) | A kind of traffic flow forecasting method average based on sliding window |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20170510 |
|
| WD01 | Invention patent application deemed withdrawn after publication |