CN112116810A - Whole road network segment travel time estimation method based on urban road checkpoint data - Google Patents
Whole road network segment travel time estimation method based on urban road checkpoint data Download PDFInfo
- Publication number
- CN112116810A CN112116810A CN202010928408.0A CN202010928408A CN112116810A CN 112116810 A CN112116810 A CN 112116810A CN 202010928408 A CN202010928408 A CN 202010928408A CN 112116810 A CN112116810 A CN 112116810A
- Authority
- CN
- China
- Prior art keywords
- travel time
- road
- data
- vehicle
- historical
- 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
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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
-
- 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
- G08G1/01—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
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Analytical Chemistry (AREA)
- Chemical & Material Sciences (AREA)
- Theoretical Computer Science (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Evolutionary Computation (AREA)
- Evolutionary Biology (AREA)
- General Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Computational Biology (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Traffic Control Systems (AREA)
Abstract
The invention provides a whole road network segment travel time estimation method based on urban road checkpoint data, which mainly comprises the following steps: 1) based on the checkpoint data, data preprocessing and screening are carried out to obtain preliminary local section travel time; 2) counting variables such as hour time period when a vehicle enters a road section, vehicle traveling times, vehicle numbers and road sections in the extracted gate data to establish a multi-dimensional tensor model; 3) performing k-means clustering analysis, and applying the result to supplement missing items of a tensor model; 4) abnormal values in the travel time are screened out, and the travel time of the outliers is estimated by utilizing a segmented interpolation method. According to the method, the vehicle travel time is extracted based on the checkpoint data, the crowds with different conditions or different driving habits are classified, missing items in the travel time are supplemented according to the result of cluster analysis, and the influence of factors such as traffic conditions and crowd driving habit differences is considered when the travel time is estimated, so that the travel time estimation of the whole road network section considering the crowd differences is realized.
Description
Technical Field
The invention relates to a method for calculating travel time of an urban road, in particular to a method for estimating travel time of a whole road network section based on urban road checkpoint data.
Background
The road section travel time is a key index in urban traffic travel information, and can be used for evaluating the service level of an urban road and reflecting the characteristics of the road such as the running condition and the like. By providing accurate travel time information, an individual driver can make better path selection, a bus management company can more effectively operate a dispatching system, a traffic management department can find out problematic places and introduce a new or modified traffic control scheme to improve the performance, and a traffic policy making department analyzes traffic demands and evaluates policy tools influencing traffic jam fees and the like.
However, travel time estimation in urban road networks is a challenging problem due to factors such as volatility in traffic demand and supply, traffic control, inherent uncertainty in random arrival and departure at signalized intersections, and the like. In recent years, advances in information collection and communication technology are transforming an area that was once data-poor into one of the most data-rich areas. More and more data collectors, such as global positioning system devices installed on taxis, wide application of high-definition bayonet devices, and the like. Large amounts of vehicle traffic data are provided on a large spatiotemporal scale and traffic information is provided for urban road networks using more complex models. However, the checkpoint data cannot provide direct information which is usually needed for analyzing models, such as traffic flow and travel time, and a pure data-driven method is needed, namely, only the data itself is used for discovering potential patterns and knowledge in the data, so that a plurality of attribute characteristics of roads or vehicles can be discovered.
However, due to the technological progress of the current day, the rapid development of the information industry and the continuous updating and upgrading of electronic equipment, the traffic information acquisition technology and equipment have great leap, and especially, the application of the future 5G technology in the fields can realize the internet of vehicles, roads and even the internet of things. Compared with the prior art, the data indexes acquired by the electronic equipment are more diversified, and the accuracy is continuously improved. Among them, the road gate camera monitoring system is one of them. However, the analysis and application of the data of the gate are mainly focused on the aspects of maintaining social security and the like, such as capturing illegal vehicles, identifying vehicles and the like. The license plate recognition data is convenient to collect, high in recognition accuracy rate and large in size, information is objective, reliable, comprehensive and continuous in recording, and compared with the traditional traffic investigation, the license plate recognition data is more efficient, so that the trip information of the vehicle can be truly reflected by using the license plate recognition data of the checkpoint, the obtained research result is more scientific and reliable, and the proposed suggestion and the formulated policy are more practical. However, the checkpoint data obtained by license plate recognition also has certain limitations, namely the encrypted license plate data cannot collect the personal social and economic attributes of the vehicle driver. Although technologies related to license plate recognition data and models have been developed to some extent in recent years, there are certain problems in terms of an effective modeling method, data sparsity, traffic condition fluctuation, and the like.
Disclosure of Invention
The purpose of the invention is as follows: aiming at the defects, the invention provides a whole road network section travel time estimation method based on urban road checkpoint data.
The technical scheme is as follows: the invention relates to a whole road network section travel time estimation method based on urban road checkpoint data, which comprises the following steps:
(1) acquiring and preprocessing a plurality of historical gate data of urban roads, wherein the historical gate data comprises gate numbers, shooting time, lanes and license plate numbers; the pretreatment specifically comprises the following steps: the method comprises the steps of firstly, renumbering M checkpoints and N vehicles in historical checkpoint data into 1-M and 1-N, then sequencing the historical checkpoint data according to vehicle numbers, and finally sequencing the historical checkpoint data corresponding to each vehicle number again according to shooting time;
(2) calculating the travel time of the road section according to the preprocessed historical checkpoint data in the step (1);
(3) establishing a tensor model based on the vehicle number, the trip time period, the road section travel time, the road section and the vehicle trip times;
(4) selecting the road section travel time, the vehicle travel times and the travel time period as variables of k-means clustering analysis, and carrying out clustering analysis;
(5) supplementing missing items of the road section travel time to the tensor model based on the result of the clustering analysis in the step (4): and performing spline interpolation on the missing item based on the clustering category where the missing item is located and the road section travel time which is the same as the corresponding trip time period and the road section.
Further, the method for calculating the travel time of the road section in the step (2) comprises the following steps:
for the nth vehicle, subtracting the shooting time in the t +1 th and the t-th preprocessed historical bayonet data to obtain the road section travel time between two bayonets corresponding to the t +1 th and the t-th preprocessed historical bayonet data, adding the travel time into the t +1 th preprocessed historical bayonet data, and deleting the 1 st preprocessed historical bayonet data corresponding to the nth vehicle to obtain new bayonet data; wherein N is 1,2, …, N, T is 1,2, …, and T is the number of the preprocessed historical bayonet data corresponding to the nth vehicle.
Further, in the step (3), the number of vehicle trips is the number of road segments passed by the vehicle, and the trip time period is the hour time period when the vehicle enters the road segments.
Further, each dimension of the tensor in the step (3) respectively represents a vehicle number, a travel time period and a road section, and the value of each item respectively represents the travel time of the nth vehicle on the p travel time period and the q road section and the travel times of the vehicle.
Has the advantages that:
(1) the invention provides a whole road network segment travel time estimation method based on urban road gate data, which utilizes urban road gate historical data with dense gate distribution to dig out the relation between the segment travel time and the gate data, and takes the factors of the difference of the driving habits of people, the traffic conditions and the like into consideration for cluster analysis, and classifies the people with different conditions or different driving habits;
(2) the invention provides an estimation method of urban road travel time missing items, wherein the missing items in the travel time are supplemented based on the results of travel time clustering analysis, the influence of factors such as traffic conditions, crowd driving habit differences and the like is considered when the travel time is estimated, and the accuracy and the adaptability of the calculated results are better, so that the travel time estimation of a whole road network section considering the crowd difference is realized.
Drawings
FIG. 1 is a flow chart of a method for estimating travel time of a full-link network segment according to the present invention;
fig. 2 is a frequency distribution histogram of a route 1,2, 3 travel time distribution, in which (a) represents the route 1, (b) represents the route 2, and (c) represents the route 3;
fig. 3 is a graph of the hourly distribution of travel times for links 1,2, 3, wherein (a) represents link 1, (b) represents link 2, and (c) represents link 3;
FIG. 4 is a tensor model;
FIG. 5 is a travel time outlier example;
FIG. 6 is a result of direct interpolation without clustering results;
fig. 7 shows the result of interpolation based on the cluster analysis result.
Detailed Description
The technical scheme of the invention is further explained by combining the attached drawings.
The method of the invention firstly calculates the travel time of the road section by using the urban road gate data, then carries out crowd characteristic analysis on the obtained travel time data, establishes a multidimensional tensor model, carries out k-means clustering, completes a road network travel time model, and carries out comparison analysis on a modeling result and the actual travel time, thereby providing corresponding travel time references for different types of drivers on the basis and having certain guiding significance for traveler path selection. Referring to fig. 1, the method includes the steps of:
(1) acquiring and preprocessing a plurality of historical gate data of urban roads, wherein the historical gate data comprises gate numbers, shooting time, lanes and license plate numbers; the pretreatment specifically comprises the following steps: the method comprises the steps of firstly, renumbering M checkpoints and N vehicles in historical checkpoint data into 1-M and 1-N, secondly, sequencing the historical checkpoint data according to vehicle numbers, and finally, sequencing the historical checkpoint data corresponding to each vehicle number again according to shooting time.
Step 1-1, data acquisition
The data used by the invention is urban road intersection data of Ningbo city main urban area 2018, 6 months and 5 days.
Step 1-2, data preprocessing
The method comprises the steps of firstly importing an original file into a database, then screening out all data in a research area according to bayonet numbers, cleaning the data, and deleting repeated data, error data and undetected data. Because the card slot number and the vehicle number in the original data are longer, in order to reduce the computer memory occupied by the file, the license plate number and the card slot number are renamed, and the fields of the vehicle Number (NEWCARID), the Lane (LANENNO), the shooting time (SNAPSHOTTIME) and the card slot Number (NEWDEVICEID) are reserved, and are sorted according to the vehicle number and the shooting time, and the preprocessed data are shown in a table 1:
TABLE 1 post-flush Bayonet data Format example
| Vehicle number | Bayonet numbering | Time of shooting | Lane |
| 1 | 33 | 2018-06-05 12:43:54.000 | 2 |
| 1 | 22 | 2018-06-05 12:44:45.000 | 2 |
| 1 | 75 | 2018-06-05 12:45:52.000 | 3 |
| 1 | 98 | 2018-06-05 12:50:07.000 | 3 |
| 1 | 144 | 2018-06-05 15:11:52.000 | 1 |
(2) For the nth vehicle, subtracting the shooting time in the t +1 th and the t th preprocessed historical bayonet data to obtain the road section travel time (unit is second) between the two bayonets corresponding to the t +1 th and the t th preprocessed historical bayonet data, adding the travel time to the t +1 th preprocessed historical bayonet data, and deleting the 1 st preprocessed historical bayonet data corresponding to the nth vehicle to obtain new bayonet data, as shown in table 2.
TABLE 2 travel time data example
| License plate number | Road travel time | Bayonet number at the beginning of road section | Bayonet numbering of |
| 1 | 51 | 33 | 22 |
| 1 | 67 | 22 | 75 |
| 1 | 255 | 75 | 98 |
| 1 | 8558 | 98 | 34 |
| 1 | 32 | 34 | 78 |
(3) And carrying out crowd characteristic analysis and space-time characteristic analysis on the travel time of some specific road sections of the urban road network, and establishing a travel time multidimensional tensor model according to the gate data. And establishing a tensor model based on variables such as vehicle numbers, road section travel time, travel time periods, road section travel times and the like.
Step 3-1, analyzing the crowd characteristics of the travel time of the urban road network
And analyzing the crowd characteristics according to the obtained travel time data. Frequency distribution histograms of the link 1,2, 3 travel time distribution are shown in (a) to (c) of fig. 2. As can be seen from the figure, the travel time of most travelers is concentrated in a certain range interval. The travel time frequency peak value of the road section 1 is concentrated on about 50s, the travel time frequency peak value of the road section 2 is concentrated on about 40s, and the travel time frequency peak value of the road section 3 is concentrated on 70-80 s. Both link 1 and link 2 show two frequency peaks, the second peak for link 1 appearing around 115s, the second frequency peak for link 2 appearing around 100s, and the second peak for link 3, although not significant, also observed around 150 s. The travel time of most of the drivers in different intervals are similar, and the travel time of the vehicle is mostly distributed in certain intervals in a centralized manner, which shows that the travel times of different drivers have certain similarity, so that the drivers with similar travel times can be considered to be subjected to cluster analysis.
The hour distribution pattern of the travel time of the links 1,2, 3 is shown in fig. 3 (a) to (c). It can be seen from the figure that the travel time of all days of the three road sections has certain similarity, the travel time of the road sections reaches the lowest value from 0 to 5, the early peak appears from 7 to 9 in the morning, the late peak appears from 17 to 19 in the evening, and the travel time of the road sections declines at other moments, and the curve change trend conforms to the theoretical knowledge of traffic engineering. As can be seen from fig. 3, the travel time of the road segment 3 at morning 8 is significantly longer than that at other times, and it is presumed that the significant increase in travel time may be caused by an increase in traffic volume at an early peak, a change in signal timing, or some other traffic management and control measures. In these figures, there are many discrete points, which are different from the travel time of most vehicles, and belong to abnormal values of the travel time, so that the travel time is corrected in the subsequent sections.
Step 3-2, establishing a multi-dimensional tensor model of road network travel time
According to the method, a tensor model is established according to variables such as vehicle numbers, travel time periods and travel times. Regarding the variable of the trip time period, the hour of the time recorded by the gate when the vehicle enters the road section is considered as the trip time period, and the hour item in the time data format 'XX year/X month/X day XX hour/XX minute' is extracted by Python to obtain the trip time period; and counting the road section trip times of each vehicle, storing the obtained result in a two-dimensional array, and correspondingly writing the road section trip times of each vehicle into a data file. As shown in fig. 4 and table 3, each dimension of the tensor represents a vehicle number, a travel period, and a link, and the value of each term represents the travel time of the nth driver (vehicle) and the number of vehicle travels on the pth travel period and the qth link.
TABLE 3 tensor model example
| License plate number | Road travel time | Trip period | Road section | Number of trips in |
| 1 | 51 | 12 | 33-22 | 5 |
| 1 | 67 | 12 | 22-75 | 5 |
| 1 | 255 | 12 | 75-98 | 5 |
| 1 | 8558 | 12 | 98-34 | 5 |
| 1 | 32 | 15 | 34-78 | 5 |
(4) And selecting the road section travel time, the road section trip times and the trip time period as variables of k-means clustering analysis, carrying out clustering analysis, processing the tensor model by using Python, substituting the processed tensor model into the established model, and setting parameters of the model and k-means clustering analysis results. From the results, the travel time can be roughly divided into three categories of fast, medium and slow, the drivers are divided into three categories from the aspects of travel time, travel time interval, travel times and the like, and the clustering center result and the clustering result are respectively shown in tables 4 and 5.
TABLE 4k mean clustering centers
| 0 | 1 | 2 | |
| 0 | -0.691239 | -0.092406 | -0.254911 |
| 1 | 1.014309 | 0.199209 | -0.233286 |
| 2 | -0.124195 | -0.303166 | 2.690035 |
Table 5k mean clustering results example
| Road travel time | Trip period | Number of trips in road section | Cluster classification | |
| 0 | -0.052083 | 0.648406 | -0.741329 | 0 |
| 1 | 0.777812 | -0.290377 | -0.446026 | 1 |
| 2 | 1.469392 | -0.290377 | -0.446026 | 1 |
| 3 | 1.773687 | 1.023919 | -0.446026 | 1 |
| 4 | -0.605346 | 1.399432 | -0.859450 | 0 |
| … | ... | ... | ... | ... |
(5) In the tensor model established, some road section travel time items have obvious deviation, as shown in table 6 and fig. 5, the travel time of the road section travel time item has great difference with the travel time of other vehicles on the same road section, and therefore the travel time item needs to be corrected. The piecewise linear interpolation in the spline interpolation method supplements the travel time missing item. During supplement, the position of the missing item in the file is firstly determined, then all data which are the same as the road section, the clustering category and the trip time period are screened out, and spline interpolation is carried out on the missing item of the travel time based on the travel time values of the data.
TABLE 6 travel time outlier example
To verify the reliability of the model, 1% of the known travel time data was randomly selected from all travel times in the study area, and two methods were used for interpolation: the first method uses direct interpolation without the clustering analysis result, and the second method estimates the interpolation based on the interpolation method, and compares the estimated value of the travel time result obtained by the two methods with the real value for analysis.
The interpolation results are shown in fig. 6 and 7, respectively, and the two graphs are subjected to regression analysis, and the images are fitted by linear regression, and the results are as follows:
1) the regression formula for interpolation according to the result of cluster analysis is:
y＝0.9695x+2.5443
wherein R is2＝0.7572。
2) The linear regression formula for directly interpolating the travel time is:
y＝0.9356x+5.9443
wherein R is2＝0.5756。
Wherein a closer coefficient β of the regression variable to 1 indicates a closer estimated value of the travel time to the true value, R2A closer to 1 indicates a higher accuracy of the travel time estimate. In practical case application, the regression coefficient may not be completely equal to 1 because the estimation result of the method has certain disadvantages and shortcomings, and the intercept is not equal to 0 may be because the estimation method has certain tendencies, such as larger overall.
As can be seen from the images and the regression results, the estimation accuracy of the method I is superior to that of the method II, the scattering points can be intuitively observed on the images to be more concentrated and converged, and the regression equation coefficients and R of the method I2The method is smaller than the method II, the fitting degree is higher, and the travel time can be estimated more accurately, so that the reliability of the method provided by the invention is verified.
Claims (4)
1. A whole road network section travel time estimation method based on urban road checkpoint data is characterized by comprising the following steps:
(1) acquiring and preprocessing a plurality of historical gate data of urban roads, wherein the historical gate data comprises gate numbers, shooting time, lanes and license plate numbers; the pretreatment specifically comprises the following steps: the method comprises the steps of firstly, renumbering M checkpoints and N vehicles in historical checkpoint data into 1-M and 1-N, then sequencing the historical checkpoint data according to vehicle numbers, and finally sequencing the historical checkpoint data corresponding to each vehicle number again according to shooting time;
(2) calculating the travel time of the road section according to the preprocessed historical checkpoint data in the step (1);
(3) establishing a tensor model based on the vehicle number, the trip time period, the road section travel time, the road section and the vehicle trip times;
(4) selecting the road section travel time, the vehicle travel times and the travel time period as variables of k-means clustering analysis, and carrying out clustering analysis;
(5) supplementing missing items of the road section travel time to the tensor model based on the result of the clustering analysis in the step (4): and performing spline interpolation on the missing item based on the clustering category where the missing item is located and the road section travel time which is the same as the corresponding trip time period and the road section.
2. The method for estimating the travel time of the road network segment based on the urban road intersection data according to claim 1, wherein the method for calculating the travel time of the road segment in the step (2) comprises the following steps:
for the nth vehicle, subtracting the shooting time in the t +1 th and the t-th preprocessed historical bayonet data to obtain the road section travel time between two bayonets corresponding to the t +1 th and the t-th preprocessed historical bayonet data, adding the travel time into the t +1 th preprocessed historical bayonet data, and deleting the 1 st preprocessed historical bayonet data corresponding to the nth vehicle to obtain new bayonet data; wherein N is 1,2, …, N, T is 1,2, …, and T is the number of the preprocessed historical bayonet data corresponding to the nth vehicle.
3. The urban road checkpoint data-based road segment travel time estimation method according to claim 2, wherein in the step (3), the number of vehicle trips is the number of road segments passed by the vehicle, and the trip time period is the hour time period when the vehicle enters the road segments.
4. The urban road intersection data-based all-road network segment travel time estimation method according to claim 3, wherein each dimension of the tensor in the step (3) represents a vehicle number, a travel time segment and a road segment, and the value of each dimension represents the travel time of the nth vehicle in the p travel time segment and the q road segment and the travel times of the nth vehicle.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010928408.0A CN112116810A (en) | 2020-09-07 | 2020-09-07 | Whole road network segment travel time estimation method based on urban road checkpoint data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010928408.0A CN112116810A (en) | 2020-09-07 | 2020-09-07 | Whole road network segment travel time estimation method based on urban road checkpoint data |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112116810A true CN112116810A (en) | 2020-12-22 |
Family
ID=73802939
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010928408.0A Pending CN112116810A (en) | 2020-09-07 | 2020-09-07 | Whole road network segment travel time estimation method based on urban road checkpoint data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112116810A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113140114A (en) * | 2021-03-09 | 2021-07-20 | 中山大学 | Vehicle travel path reconstruction method based on travel time estimation |
| CN113487872A (en) * | 2021-09-07 | 2021-10-08 | 南通飞旋智能科技有限公司 | Bus transit time prediction method based on big data and artificial intelligence |
| CN116386336A (en) * | 2023-05-29 | 2023-07-04 | 四川国蓝中天环境科技集团有限公司 | Road network traffic flow robust calculation method and system based on bayonet license plate data |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105740401A (en) * | 2016-01-28 | 2016-07-06 | 北京理工大学 | Individual behavior and group interest-based interest place recommendation method and device |
| CN106997389A (en) * | 2017-03-30 | 2017-08-01 | 浙江鸿程计算机系统有限公司 | Recommend method in a kind of tourist attractions based on many data sets and collaboration tensor resolution |
| CN109308804A (en) * | 2018-08-08 | 2019-02-05 | 北京航空航天大学 | Hourage estimation method based on tensor resolution |
| CN110674111A (en) * | 2019-09-18 | 2020-01-10 | 北京航空航天大学 | Tensor decomposition-based vehicle missing travel time filling method |
| CN110689181A (en) * | 2019-09-18 | 2020-01-14 | 北京航空航天大学 | Travel time judgment method based on collaborative tensor decomposition |
| CN111311904A (en) * | 2020-01-17 | 2020-06-19 | 东南大学 | Traffic state estimation method based on floating car data weighted tensor reconstruction |
-
2020
- 2020-09-07 CN CN202010928408.0A patent/CN112116810A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105740401A (en) * | 2016-01-28 | 2016-07-06 | 北京理工大学 | Individual behavior and group interest-based interest place recommendation method and device |
| CN106997389A (en) * | 2017-03-30 | 2017-08-01 | 浙江鸿程计算机系统有限公司 | Recommend method in a kind of tourist attractions based on many data sets and collaboration tensor resolution |
| CN109308804A (en) * | 2018-08-08 | 2019-02-05 | 北京航空航天大学 | Hourage estimation method based on tensor resolution |
| CN110674111A (en) * | 2019-09-18 | 2020-01-10 | 北京航空航天大学 | Tensor decomposition-based vehicle missing travel time filling method |
| CN110689181A (en) * | 2019-09-18 | 2020-01-14 | 北京航空航天大学 | Travel time judgment method based on collaborative tensor decomposition |
| CN111311904A (en) * | 2020-01-17 | 2020-06-19 | 东南大学 | Traffic state estimation method based on floating car data weighted tensor reconstruction |
Non-Patent Citations (2)
| Title |
|---|
| 邱淳风 等: "基于支持向量回归的行程时间预测算法", 《计算机时代》 * |
| 陶莎 等: "基于无线探针数据的区间车速估计方法与分析", 《武汉理工大学学报（交通科学与工程版）》 * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113140114A (en) * | 2021-03-09 | 2021-07-20 | 中山大学 | Vehicle travel path reconstruction method based on travel time estimation |
| CN113140114B (en) * | 2021-03-09 | 2022-07-29 | 中山大学 | Vehicle travel path reconstruction method based on travel time estimation |
| CN113487872A (en) * | 2021-09-07 | 2021-10-08 | 南通飞旋智能科技有限公司 | Bus transit time prediction method based on big data and artificial intelligence |
| CN113487872B (en) * | 2021-09-07 | 2021-11-16 | 南通飞旋智能科技有限公司 | Bus transit time prediction method based on big data and artificial intelligence |
| CN116386336A (en) * | 2023-05-29 | 2023-07-04 | 四川国蓝中天环境科技集团有限公司 | Road network traffic flow robust calculation method and system based on bayonet license plate data |
| CN116386336B (en) * | 2023-05-29 | 2023-08-08 | 四川国蓝中天环境科技集团有限公司 | Road network traffic flow robust calculation method and system based on bayonet license plate data |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109871876B (en) | Expressway road condition identification and prediction method based on floating car data | |
| CN112116810A (en) | Whole road network segment travel time estimation method based on urban road checkpoint data | |
| Mazimpaka et al. | Trajectory data mining: A review of methods and applications | |
| CN104318324B (en) | Shuttle Bus website and route planning method based on taxi GPS records | |
| Saha et al. | Pattern recognition using clustering analysis to support transportation system management, operations, and modeling | |
| CN113159364A (en) | Passenger flow prediction method and system for large-scale traffic station | |
| CN113011815A (en) | Truck stop point extraction method and travel characteristic determination method and device | |
| CN114463972A (en) | Road section interval traffic analysis and prediction method based on ETC portal communication data | |
| Chen et al. | Extracting bus transit boarding stop information using smart card transaction data | |
| Turkmen et al. | Intermittent demand forecasting with deep renewal processes | |
| Chen et al. | An analysis of movement patterns between zones using taxi GPS data | |
| Lee et al. | Travel pattern-based bus trip origin-destination estimation using smart card data | |
| CN108681741B (en) | Subway commuting crowd information fusion method based on IC card and resident survey data | |
| CN108053646B (en) | Traffic characteristic obtaining method, traffic characteristic prediction method and traffic characteristic prediction system based on time sensitive characteristics | |
| CN112101132B (en) | Traffic condition prediction method based on graph embedding model and metric learning | |
| CN110020666B (en) | Public transport advertisement putting method and system based on passenger behavior mode | |
| Liu et al. | Method for identifying truck traffic site clustering using weigh-in-motion (WIM) data | |
| CN110459050B (en) | Short-term bus passenger flow prediction method based on mixed decision tree | |
| Salamanis et al. | Evaluating the effect of time series segmentation on STARIMA-based traffic prediction model | |
| CN112559909B (en) | Business area discovery method based on GCN embedded spatial clustering model | |
| Pavlyuk et al. | Spatiotemporal dynamics of public transport demand: a case study of Riga | |
| Sreedhar et al. | Road Traffic Accident Analysis and Visualization of Accident Prone Areas | |
| Lee et al. | Time-Series Data and Analysis Software of Connected Vehicles. | |
| Giraldo-Forero et al. | A Spatiotemporal Analysis of Taxis Demand: A Case Study in the Manizales City | |
| CN116665342B (en) | New energy automobile driving behavior analysis method, system and equipment |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20201222 |