CN108648445B - Dynamic traffic situation prediction method based on traffic big data - Google Patents
Dynamic traffic situation prediction method based on traffic big data Download PDFInfo
- Publication number
- CN108648445B CN108648445B CN201810357889.7A CN201810357889A CN108648445B CN 108648445 B CN108648445 B CN 108648445B CN 201810357889 A CN201810357889 A CN 201810357889A CN 108648445 B CN108648445 B CN 108648445B
- Authority
- CN
- China
- Prior art keywords
- data
- pca
- situation
- traffic
- cent
- 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
Images
Classifications
-
- 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
- G08G1/0129—Traffic data processing for creating historical data or processing based on historical data
-
- 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
Landscapes
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Traffic Control Systems (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a dynamic traffic situation prediction method based on traffic big data.
Description
Technical Field
The invention relates to the technical field of traffic big data situation prediction, in particular to a dynamic traffic situation prediction method based on traffic big data, which is high in accuracy and good in stability.
Background
At present, with the rapid increase of the number of motor vehicles in China, urban traffic congestion becomes a problem commonly faced by various large and medium cities in China, delay time and environmental pollution caused by congestion bring huge economic loss to the society, and the real-time traffic situation published by the prior art is the real-time traffic situation of the current road section, and the prediction result of the traffic situation in a period of time in the future cannot be displayed. However, in real life, more and more users want to know the traffic conditions of some road sections or some areas in advance for reasonably arranging the travel route in advance, and therefore the existing real-time traffic situation cannot meet the route planning requirements of the users. Therefore, the research on the traffic situation can provide important basis for the selection of the paths of travelers and the dispersion of traffic by traffic managers, and the accurate prediction of the traffic situation becomes a problem to be solved urgently at present.
In the existing traffic data collection method, a license plate recognition technology is a new urban road traffic information collection technology, vehicle license plates are captured, recognized and recorded through a bayonet camera arranged at an urban road intersection, the license plate recognition technology has the advantages of high precision, large sample quantity, wide vehicle coverage range and the like in the aspect of traffic data acquisition, historically accumulated bayonet data form traffic big data, and the average speed of a vehicle interval is used for calculating the speed of a road section travel, so the traffic situation discrimination method based on the bayonet system license plate recognition has good academic research significance.
Scholars at home and abroad make a great deal of research on traffic situation prediction algorithms. The learners select the traffic flow, the occupancy, the traffic speed and the traffic density of the road section as parameters and predict the probability of various traffic situations by using a Bayesian algorithm. The learners also use the parameters such as flow, speed, occupancy, average lane flow and the like to obtain the weight corresponding to each parameter by adopting the maximum entropy training model, thereby predicting the traffic situation. There is also a traffic flow state fuzzy inference method based on neural network, the algorithms all consider a plurality of parameters, and parameter combinations corresponding to traffic situations are obtained through training. However, the excessive parameter input causes great limitation in practical application. And different basic situation tables are established by distinguishing working days, non-working days and weather states through historical road section speed data, and the traffic situation of the whole day is predicted only according to the current date and the weather. The method has low prediction accuracy due to too large uncertain factors.
Disclosure of Invention
The invention aims to overcome the defect that the accuracy and the stability of the traffic situation prediction method in the prior art can not meet the traffic demand, and provides a dynamic traffic situation prediction method based on traffic big data, which is high in accuracy and good in stability.
In order to achieve the purpose, the invention adopts the following technical scheme:
a dynamic traffic situation prediction method based on traffic big data is characterized by comprising the following steps:
(1-1) collecting traffic data information of regional road sections;
(1-2) acquiring n traffic data information of the regional road section within M days before the current timeSampling data at sampling intervals T1；
(1-3) standardizing the sampling data, analyzing the standardized sampling data by using a principal component analysis method to obtain characteristic data pca _ v of the sampling data, converting the characteristic data pca _ v into a range between [ -a, a ], and converting the characteristic data pca _ v into [0, 1 ];
(1-4) dividing the situation;
(1-5) selecting pca _ v data with the same time length S every day within M days before the current moment, wherein the total number of the pca _ v data is M
Line data pca _ vj1Setting the clustering radius as r, clustering by adopting a K-Means clustering method to obtain a clustering result as E class, and the clustering center as cent _ vi1Storing the data into a historical learning library cent _ v;
(1-6) sampling data of the traffic data information with the time length S away from the current moment is obtained, and the steps (1-2), (1-3) and (1-4) are repeated to obtain one-dimensional data pca _ vc;
(1-7) selecting the time length of pca _ vc and the historical learning base cent _ v as
Calculating the distance dis and cosine distance dis _ cos between pca _ vc and the historical learning base cent _ v; and making a traffic situation prediction;
(1-8) updating the historical learning library cent _ v.
The existing prediction method can not meet the traffic requirements in the aspects of accuracy and stability, and the invention adopts a rolling dynamic optimization thought to continuously update the model according to real time and history in the process of predicting the situation, so that the predicted situation is more in line with the reality.
Preferably, the step (1-3) comprises the following specific steps:
forming an n multiplied by 3 data matrix by using the sampled data, wherein elements in the data matrix are represented by xij;
(2-1) the normalization process is performed using the following formula:
i＝1，2，...，n；j＝1，2，3；
wherein the content of the first and second substances,
obtaining a standardized data matrix;
(2-2) setting the sample correlation coefficient matrix to R,calculating each element R in the sample correlation coefficient matrix Rij：
(2-3)
wherein the new data F of the samplei1＝a11xi1+a12xi2+a13xi3，[a11，a12，a13]Calculating a feature vector obtained by the correlation coefficient matrix R;
(2-4) adding FijConversion to [ -a, a [ -a]，
(2-5) adding FijConversion to [0, 1]，Fij＝(Fij+a)/2a。
Preferably, the sampling data comprises the actual travel speed v, the inverse number 1/q of the traffic flow at the intersection downstream of the road section and the standard time difference std _ time of the headway downstream of the road section.
Preferably, the step (1-7) further comprises the steps of:
if dis < r exists, the data in the last half hour in the historical learning library cent _ v which meets the condition that dis < r and has the smallest cosine distance dis _ cos are taken as a prediction situation result;
if dis > r, an alarm is given, and if pca _ vc is severely missing data, no prediction is given.
Preferably, the steps (1-4) comprise the following specific steps:
divide into situation 1 when pca _ v is at [0, 0.1), divide into situation 2 when pca _ v is at [0.1, 0.2), divide into situation 3 when pca _ v is at [0.2, 0.3), divide into situation 4 when pca _ v is at [0.3, 0.4), divide into situation 5 when pca _ v is at [0.4, 0.5), divide into situation 6 when pca _ v is at [0.5, 0.6), divide into situation 7 when pca _ v is at [0.6, 0.7), divide into situation 8 when pca _ v is at [0.7, 0.8), divide into situation 9 when pca _ v is at [0.8, 0.9), divide into situation 10 when pca _ v is at [0.9, 1 ].
Preferably, the steps (1-8) comprise the following specific steps:
calculating the distance D1 between the data in the S time length before the current time and the data in each corresponding S time length in the historical learning base cent _ v, selecting the smallest D1, and classifying the current data into the historical learning base cent _ v corresponding to the smallest D1;
and updating the clustering center cent _ v of the historical learning base cent _ vi；
cent_vi＝(cent_vi+ pca-vc)/2, where cent _ viIs a cluster center in cent _ v.
Preferably, M is 20 to 40 days, T15 minutes or 10 minutes, and S is 1.5 hours.
Therefore, the invention has the following beneficial effects: and by adopting a rolling dynamic optimization idea, the model is continuously updated according to real time and history in the process of predicting the situation, so that the predicted situation is more practical.
Drawings
FIG. 1 is a flow chart of the present invention;
FIG. 2 is a schematic diagram of an experimental road network area according to the present invention;
fig. 3 is a schematic diagram of a traffic situation rolling prediction method according to the present invention.
FIG. 4 is a graph comparing the predicted value and the actual value of pca _ vc for 5 minutes of rolling prediction from 6 o 'clock to 22 o' clock during a certain segment of the day according to the present invention;
FIG. 5 is a graph comparing the predicted value and the actual value of the half-hour pca _ vc predicted by 6 o 'clock to 22 o' clock rolling in the daytime for a certain segment of the present invention;
FIG. 6 is a schematic diagram illustrating comparison between the true value and the predicted value of the situation of a road segment in the experimental road network.
Detailed Description
The invention is further described with reference to the following figures and detailed description.
As shown in fig. 1 and fig. 2, the invention provides a dynamic traffic situation prediction method based on traffic big data, taking a local area network of a cauda bridge as an example, selecting a cauda bridge road from a west-going duyan road, a north-going Yumin road and a south-going quxian road, wherein the cauda bridge road comprises 7 road sections;
the method for dynamically predicting the road section traffic situation of the area comprises the following steps:
(1-1) collecting traffic data information of regional road sections;
(1-2) acquiring 5760 sampling data from 6 am to 10 pm in the daytime of one month before the current moment of the regional road section, wherein the sampling data comprise an actual travel speed v, the reciprocal 1/q of the traffic flow of a downstream intersection of the road section and the standard time difference std _ time of the time interval of the downstream locomotive of the road section, and the sampling interval is 5 minutes;
(1-3) standardizing the sampling data, analyzing the standardized sampling data by using a principal component analysis method to obtain characteristic data pca _ v of the sampling data, converting the characteristic data pca _ v into a range between [ -a, a ], and converting the characteristic data pca _ v into [0, 1 ];
forming an n x 3 data matrix from the sampled data, the elements in the data matrix being xijRepresents;
(1-3-1) the normalization process was performed using the following formula:
i＝1，2，...，n；j＝1，2，3；
wherein the content of the first and second substances,
obtaining a standardized data matrix;
(1-3-2) setting the sample correlation coefficient matrix to R,
calculating each element R in the sample correlation coefficient matrix Rij：
(1-3-3)
wherein the new data F of the samplei1＝a11xi1+a12xi2+a13xi3，[a11，a12，a13]Calculating a feature vector obtained by the correlation coefficient matrix R;
(1-3-4) adding FijConversion to [ -a, a [ -a]，
(1-3-5) adding FijConversion to [0, 1]，Fij＝(Fij+a)/2a。
(1-4) dividing the situation;
divide into situation 1 when pca _ v is at [0, 0.1), divide into situation 2 when pca _ v is at [0.1, 0.2), divide into situation 3 when pca _ v is at [0.2, 0.3), divide into situation 4 when pca _ v is at [0.3, 0.4), divide into situation 5 when pca _ v is at [0.4, 0.5), divide into situation 6 when pca _ v is at [0.5, 0.6), divide into situation 7 when pca _ v is at [0.6, 0.7), divide into situation 8 when pca _ v is at [0.7, 0.8), divide into situation 9 when pca _ v is at [0.8, 0.9), divide into situation 10 when pca _ v is at [0.9, 1 ]. The smaller the situation value is, the more the traffic is congested, and the original data and situation data of part of the road sections are as shown in table 1:
(1-5) selecting 1 half-hour of pca _ v data at the same time in 30 days, and totally 30 groups of 18 rows of data pca _ vj1Selecting a K-Means cluster pca _ v with a cluster radius rjAfter clustering, the cluster is divided into E classes, and the clustering center is cent _ vi1And storing the data into a historical learning base cent _ v, wherein the clustered partial cent _ v data are shown in a table 2:
(1-6) sampling data of traffic data information within one hour away from the current moment is taken, and the steps (1-2), (1-3) and (1-4) are repeated to obtain current one-dimensional data pca _ vc;
(1-7) calculating the distance dis and the cosine distance dis _ cos between the pca _ vc and the historical learning base cent _ v (12 pieces of data in one hour corresponding to the same moment are selected), if dis < r exists, predicting, and if dis < r exists, the data in the last half hour in a reference array cent _ v which meets the condition that dis < r and the cosine distance dis _ cos is minimum are the predicted situation result, if dis > r, giving an alarm, and if the pca _ vc lacks data seriously, giving no prediction, wherein the distance and included angle calculation method comprises the following steps:
(1-8) calculating the distance D1 between the data within 1 half hour before the current time and the data of each corresponding S time length in the historical learning base cent _ v, selecting the smallest D1, and classifying the current data into the historical learning base cent _ v corresponding to the smallest D1;
updating a clustering center cent _ vi of a historical learning base cent _ v;
cent_vi＝(cent_vi+ pca _ vc)/2, where cent _ viIs a cluster center in cent _ v.
(1-9) in fig. 3, it is assumed that one period is [ t-Tr t + Tr ], then the next period is [ t + Tr t +2Tr ], and one period prediction is finished, and the step (1-6) is skipped to perform the next period prediction.
By adopting the rolling prediction method, the predicted value and the real value of the pca _ vc are rolled and predicted for 5 minutes from 6 points to 22 points in the day of a certain path, for example, as shown in fig. 4 (in the figure, the asterisks represent the predicted value and the real value of the line), the predicted value and the real value of the pca _ vc are rolled and predicted for half an hour from 6 points to 22 points in the day, for example, as shown in fig. 5 (in the figure, the asterisks represent the predicted value and the real value of the line), and the correlation between the situation predicted value and the real value reaches 0.77.
Fig. 6 is a schematic diagram comparing the real value and the predicted value of the situation of a certain road segment in the experimental road network, the thin line represents the real value, the thick line represents the predicted value, the situation of the change of the certain road segment along with time can be seen from fig. 6, the trend of the real value and the predicted value is consistent, and the correlation between the situation predicted value and the actual value reaches 0.77.
The above examples are provided only for illustrating the present invention and are not intended to limit the present invention. Changes, modifications, etc. to the above-described embodiments are intended to fall within the scope of the claims of the present invention as long as they are in accordance with the technical spirit of the present invention.
Claims (7)
1. A dynamic traffic situation prediction method based on traffic big data is characterized by comprising the following steps:
(1-1) collecting traffic data information of regional road sections;
(1-2) acquiring n sampling data of traffic data information within M days before the current time of the distance of a regional road section, wherein the sampling interval is T1；
(1-3) standardizing the sampling data, analyzing the standardized sampling data by using a principal component analysis method to obtain characteristic data pca _ v of the sampling data, converting the characteristic data pca _ v into a range between [ -a, a ], and converting the characteristic data pca _ v into [0, 1 ];
(1-4) dividing the situation;
(1-5) selecting pca _ v data with the same time length S every day within M days before the current moment, wherein the total number of the pca _ v data is M
Line data pca _ vj1Setting the clustering radius as r, clustering by adopting a K-Means clustering method to obtain a clustering result as E class, and the clustering center as cent _ vi1Storing the data into a historical learning library cent _ v;
(1-6) sampling data of the traffic data information with the time length S away from the current moment is obtained, and the steps (1-2), (1-3) and (1-4) are repeated to obtain one-dimensional data pca _ vc;
(1-7) selecting the time length of pca _ vc and the historical learning base cent _ v asCalculating the distance dis and cosine distance dis _ cos between pca _ vc and the historical learning base cent _ v; and making a traffic situation prediction;
(1-8) updating the historical learning library cent _ v.
2. The dynamic traffic situation prediction method based on traffic big data according to claim 1, characterized in that the step (1-3) comprises the following specific steps:
forming an n x 3 data matrix from the sampled data, the elements in the data matrix being xijRepresents;
(2-1) the normalization process is performed using the following formula:
i＝1，2，...，n；j＝1，2，3；
wherein the content of the first and second substances,
obtaining a standardized data matrix;
(2-2) setting the sample correlation coefficient matrix to R,
calculating each element R in the sample correlation coefficient matrix Rij：
(2-3)
wherein the new data F of the samplei1＝a11xi1+a12xi2+a13xi3，[a11，a12，a13]Calculating a feature vector obtained by the correlation coefficient matrix R;
(2-4) adding FijConversion to [ -a, a [ -a]，
(2-5) adding FijConversion to [0, 1]，Fij＝(Fij+a)/2a。
3. The dynamic traffic situation prediction method based on the traffic big data as claimed in claim 1, wherein the sampled data comprises an actual travel speed v, an inverse number 1/q of traffic flow at a junction downstream of the road section and a standard head-time difference std _ time downstream of the road section.
4. The dynamic traffic situation prediction method based on traffic big data according to claim 1, characterized in that the step (1-7) further comprises the steps of:
if dis < r exists, the data in the last half hour in the historical learning library cent _ v which meets the condition that dis < r and has the smallest cosine distance dis _ cos are taken as a prediction situation result;
if dis > r, an alarm is given, and if pca _ vc is severely missing data, no prediction is given.
5. The dynamic traffic situation prediction method based on traffic big data according to claim 1, characterized in that the step (1-4) comprises the following specific steps:
divide into situation 1 when pca _ v is at [0, 0.1), divide into situation 2 when pca _ v is at [0.1, 0.2), divide into situation 3 when pca _ v is at [0.2, 0.3), divide into situation 4 when pca _ v is at [0.3, 0.4), divide into situation 5 when pca _ v is at [0.4, 0.5), divide into situation 6 when pca _ v is at [0.5, 0.6), divide into situation 7 when pca _ v is at [0.6, 0.7), divide into situation 8 when pca _ v is at [0.7, 0.8), divide into situation 9 when pca _ v is at [0.8, 0.9), divide into situation 10 when pca _ v is at [0.9, 1 ].
6. The dynamic traffic situation prediction method based on traffic big data according to claim 1, characterized in that the steps (1-8) comprise the following specific steps:
calculating the distance D1 between the data in the S time length before the current time and the data in each corresponding S time length in the historical learning base cent _ v, selecting the smallest D1, and classifying the current data into the historical learning base cent _ v corresponding to the smallest D1;
and updating the clustering center cent _ v of the historical learning base cent _ vi；
cent_vi＝(cent_vi+ pca _ vc)/2, where cent _ viIs a cluster center in cent _ v.
7. The dynamic traffic situation prediction method based on traffic big data according to claim 1, 2, 3, 4, 5 or 6, wherein M is 20 days to 40 days, T15 minutes or 10 minutes, and S is 1.5 hours.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810357889.7A CN108648445B (en) | 2018-04-19 | 2018-04-19 | Dynamic traffic situation prediction method based on traffic big data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810357889.7A CN108648445B (en) | 2018-04-19 | 2018-04-19 | Dynamic traffic situation prediction method based on traffic big data |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108648445A CN108648445A (en) | 2018-10-12 |
| CN108648445B true CN108648445B (en) | 2020-02-21 |
Family
ID=63746700
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810357889.7A Active CN108648445B (en) | 2018-04-19 | 2018-04-19 | Dynamic traffic situation prediction method based on traffic big data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108648445B (en) |
Families Citing this family (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112907267A (en) * | 2019-12-03 | 2021-06-04 | 顺丰科技有限公司 | Method and device for predicting cargo quantity, computer equipment and storage medium |
| CN111523562B (en) * | 2020-03-20 | 2021-06-08 | 浙江大学 | Commuting mode vehicle identification method based on license plate identification data |
| CN112069376A (en) * | 2020-08-27 | 2020-12-11 | 武汉理工大学 | Data processing method, system and storage medium for traffic information visualization |
| CN112598169B (en) * | 2020-12-17 | 2023-03-24 | 广东南方通信建设有限公司 | Traffic operation situation assessment method, system and device |
| CN113030443A (en) * | 2021-02-26 | 2021-06-25 | 上海伽易信息技术有限公司 | Intelligent monitoring method and judgment model for oil of metro vehicle based on dynamic self-adaptive trend analysis |
| CN115440045B (en) * | 2021-08-19 | 2023-12-22 | 佛山市城市规划设计研究院 | Method for improving real-time accuracy of traffic navigation |
| CN114023073B (en) * | 2022-01-06 | 2022-04-19 | 南京感动科技有限公司 | Expressway congestion prediction method based on vehicle behavior analysis |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103810849A (en) * | 2012-11-15 | 2014-05-21 | 北京掌城科技有限公司 | Traffic flow change trend extraction method based on floating car data |
| CN104504897A (en) * | 2014-09-28 | 2015-04-08 | 北京工业大学 | Intersection traffic flow characteristic analysis and vehicle moving prediction method based on trajectory data |
| CN104778837A (en) * | 2015-04-14 | 2015-07-15 | 吉林大学 | Multi-time scale forecasting method for road traffic running situation |
| CN107154150A (en) * | 2017-07-25 | 2017-09-12 | 北京航空航天大学 | A kind of traffic flow forecasting method clustered based on road with double-layer double-direction LSTM |
| CN107451599A (en) * | 2017-06-28 | 2017-12-08 | 青岛科技大学 | A kind of traffic behavior Forecasting Methodology of the collective model based on machine learning |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9368027B2 (en) * | 2013-11-01 | 2016-06-14 | Here Global B.V. | Traffic data simulator |
-
2018
- 2018-04-19 CN CN201810357889.7A patent/CN108648445B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103810849A (en) * | 2012-11-15 | 2014-05-21 | 北京掌城科技有限公司 | Traffic flow change trend extraction method based on floating car data |
| CN104504897A (en) * | 2014-09-28 | 2015-04-08 | 北京工业大学 | Intersection traffic flow characteristic analysis and vehicle moving prediction method based on trajectory data |
| CN104778837A (en) * | 2015-04-14 | 2015-07-15 | 吉林大学 | Multi-time scale forecasting method for road traffic running situation |
| CN107451599A (en) * | 2017-06-28 | 2017-12-08 | 青岛科技大学 | A kind of traffic behavior Forecasting Methodology of the collective model based on machine learning |
| CN107154150A (en) * | 2017-07-25 | 2017-09-12 | 北京航空航天大学 | A kind of traffic flow forecasting method clustered based on road with double-layer double-direction LSTM |
Non-Patent Citations (1)
| Title |
|---|
| 基于非参数回归的短时交通流量预测方法研究;范鲁明;《中国优秀硕士学位论文全文数据库》;20090430;全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108648445A (en) | 2018-10-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108648445B (en) | Dynamic traffic situation prediction method based on traffic big data | |
| CN109871876B (en) | Expressway road condition identification and prediction method based on floating car data | |
| CN108346292B (en) | Urban expressway real-time traffic index calculation method based on checkpoint data | |
| CN104318324B (en) | Shuttle Bus website and route planning method based on taxi GPS records | |
| CN106228808B (en) | City expressway travel time prediction method based on Floating Car space-time grid data | |
| Li et al. | Identifying important variables for predicting travel time of freeway with non-recurrent congestion with neural networks | |
| CN114783183B (en) | Traffic situation algorithm-based monitoring method and system | |
| CN107610469A (en) | A kind of day dimension regional traffic index forecasting method for considering multifactor impact | |
| CN110390349A (en) | Bus passenger flow volume based on XGBoost model predicts modeling method | |
| CN113159364A (en) | Passenger flow prediction method and system for large-scale traffic station | |
| Loulizi et al. | Steady‐State Car‐Following Time Gaps: An Empirical Study Using Naturalistic Driving Data | |
| CN110400462B (en) | Track traffic passenger flow monitoring and early warning method and system based on fuzzy theory | |
| CN109376906B (en) | Travel time prediction method and system based on multi-dimensional trajectory and electronic equipment | |
| CN112016735B (en) | Patrol route planning method and system based on traffic violation hotspot prediction and readable storage medium | |
| Bachechi et al. | Traffic analysis in a smart city | |
| CN114463972A (en) | Road section interval traffic analysis and prediction method based on ETC portal communication data | |
| CN113537569B (en) | Short-term bus passenger flow prediction method and system based on weight stacking decision tree | |
| CN113112076B (en) | Passenger flow distribution mode discovery and prediction method | |
| CN114091581A (en) | Vehicle operation behavior type identification method based on sparse track | |
| CN114463978B (en) | Data monitoring method based on track traffic information processing terminal | |
| Loubesi et al. | Road trafficking description and short term travel time forecasting, with a classification method | |
| CN112101132B (en) | Traffic condition prediction method based on graph embedding model and metric learning | |
| CN114139984A (en) | Urban traffic accident risk prediction method based on flow and accident collaborative perception | |
| CN115035715B (en) | Expressway flow prediction method based on decision tree and multi-element auxiliary information | |
| CN111275961A (en) | Urban traffic running state feature calculation method based on floating car data |
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 | ||
| CP01 | Change in the name or title of a patent holder | ||
| CP01 | Change in the name or title of a patent holder |
Address after: 310053 23-25, 2 building, 352 BINKANG Road, Binjiang District, Hangzhou, Zhejiang. Patentee after: Zhejiang zhongkong Information Industry Co.,Ltd. Address before: 310053 23-25, 2 building, 352 BINKANG Road, Binjiang District, Hangzhou, Zhejiang. Patentee before: ZHEJIANG SUPCON INFORMATION TECHNOLOGY Co.,Ltd. |