CN102074124A - Dynamic bus arrival time prediction method based on support vector machine (SVM) and H-infinity filtering - Google Patents
Dynamic bus arrival time prediction method based on support vector machine (SVM) and H-infinity filtering Download PDFInfo
- Publication number
- CN102074124A CN102074124A CN 201110028817 CN201110028817A CN102074124A CN 102074124 A CN102074124 A CN 102074124A CN 201110028817 CN201110028817 CN 201110028817 CN 201110028817 A CN201110028817 A CN 201110028817A CN 102074124 A CN102074124 A CN 102074124A
- Authority
- CN
- China
- Prior art keywords
- time
- svm
- arrival time
- bus
- website
- 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.)
- Granted
Links
Images
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Traffic Control Systems (AREA)
Abstract
The invention discloses a dynamic bus arrival time prediction method based on a support vector machine (SVM) and H-infinity filtering, which comprises the following two parts: 1, based on an SVM process, constructing an SVM model by using a historical time database recording the arrival times of a bus at all station, determining the value of an input variable, and predicting the running time between two adjacent stations; and 2, based on an H-infinity filtering concept, predicting the arrival time of the bus at each downstream station by combining the real-time bus running information and the running time between two adjacent stations. In the method, bus arrival time historical data and bus running real-time information are taken into comprehensive consideration to improve the accuracy of the arrival time prediction; in the dynamic prediction part, the H-infinity filtering concept is introduced for the first time, no noise statistic characteristic assumption is made, and the robustness of the prediction on the arrival time is increased; and the SVM is constructed in an off-line mode, and the effectiveness of the arrival time prediction is solved. The entire method improves the accuracy, robustness and effectiveness of the arrival time prediction.
Description
Technical field:
The present invention relates to a kind of based on SVM and H
∞The dynamic public transport arrival time Forecasting Methodology of filtering belongs to the intelligent transportation system field.
Technical background:
As the informationalized gordian technique of public transit system, real-time accurate public transport arrival time prediction not only can convenience traveler selects circuit, alleviation by bus to wait for the anxious mood of traveler, and provide strong foundation for public traffic management department science, efficient, rational management public transit vehicle, thereby greatly enrich the public transport transportation service content, establish public transport transportation favorable image, can attract more traveler to select the bus trip mode, and then alleviate the urban traffic pressure to a certain extent.
Existing public transport arrival time Forecasting Methodology has following several: the Forecasting Methodology based on historical data is prerequisite by the variation of hypothesis traffic circulation mold cycle, actual travel situation around historical travel situations fluctuation within a narrow range; Journey time is identical with the current time journey time constantly to suppose next based on the real time data Forecasting Methodology; Time series models are by seeking inherent mathematical law that exists between historical data and then the value that dopes non-independent variable; Variable decay Forecasting Methodology is a variable with a plurality of key factors, obtains stroke function between the station, sets up the mathematical prediction model that bus arrives the downstream website time; Imitate the mode of learning of human brain based on the Forecasting Methodology of artificial bionic intelligent algorithm, realize by two stages of training and testing; The Kalman wave filter is an optimization autoregression data processing algorithm, and the mode that this method utilization constantly approaches obtains higher forecast precision.Can't reflect dynamically changeable traffic behavior based on the Forecasting Methodology of historical data, time series models, variable attenuation model can't reflect Real-time Traffic Information equally, are subjected to ectocine bigger based on the Forecasting Methodology of real time data.Because its training time length can't be used for real-time estimate, the Kalman wave filter is because it is strong excessively to noise constraints based on the artificial bionic intelligent algorithm, and the actual prediction performance reduces.
Comprehensive above the analysis, the present invention proposes a kind of based on SVM and H
∞The dynamic bus arrival time forecasting methods of filtering.This method synthesis has been considered bus arrival time history data and bus running real-time information, has improved the accuracy of arrival time prediction; The performance prediction part is introduced H first
∞Filtering thought is not made any hypothesis to noise statistics, has strengthened the robustness of arrival time prediction; The SVM model construction adopts offline mode, has solved the effective problem of arrival time prediction.Entire method has improved accuracy, robustness and the actual effect of arrival time prediction.
Summary of the invention:
The object of the present invention is to provide a kind of based on SVM and H
∞The dynamic bus arrival time forecasting methods of filtering.This method is divided into two parts: static prediction part and performance prediction part.Static prediction partly is based on historical data base, adopt between the SVM method prediction adjacent sites working time; Performance prediction partly is based on H
∞Filtering thought, according to the real-time vehicle operation information and in conjunction with between the adjacent sites working time, the performance prediction public transit vehicle arrives the time of each website.
A kind of based on SVM and H
∞The dynamic bus arrival time forecasting methods of filtering may further comprise the steps:
(1) arrives the historical time database of each website according to bus, raw data is carried out pre-service, preprocessed data is put in order the data set that obtains meeting the preset recording form;
(2) by the method for statistical study the data set that step (1) obtains is classified, obtain flat peak class data set on working day, working day peak class data set, weekend flat peak class data set and weekend peak class data set;
(3) choose the input variable of SVM model construction, adopt the SVM method that the data set that classification in the step (2) obtains is carried out SVM model construction and test, set the measuring accuracy threshold value of SVM model, if the measuring accuracy of SVM model reaches the threshold value requirement, the SVM model construction is finished, otherwise rebuilds the SVM model; Described input variable comprises weather, highway section, time period and date;
Behind the historical time database update, repeating step (1), (2), (3) rebuild and test the SVM model;
(4) determine the numerical value of input variable, make up the SVM model of finishing according to step (3), obtain on the public bus network between the adjacent sites working time;
(5) according to filter state equation and observation equation, the real time execution information that automatic station name announcing system provides on working time on the public bus network that integrating step (4) obtains between the adjacent sites and the bus adopts H
∞Filtering method obtains the time that bus arrives each website of downstream, realizes dynamic bus arrival time prediction; Described real time execution information comprises that vehicle reaches website time, public transport operation line number, public transport arrival station period.
The pre-service of the raw data described in the step (1) comprises two stages: the processing stage of obliterated data and the journey time calculation stages; Be according to same vehicle, two record fields of same line raw data to be sorted the processing stage of obliterated data, whether continuous by detecting the website number field, whether judgment data loses; If lose the record strip number greater than 3, comprise 3, then abandon all station datas that comprise this record; If losing the record strip number is less than 3, then pass through the length computation polishing in adjacent upstream website and downstream website average running speed and current highway section; The journey time calculation stages is to utilize adjacent two website arrival times to subtract each other to obtain.
Data record format described in the step (1) comprises a plurality of fields, as Link Travel Time, weather, highway section, flat peak time section, rush hour section, working day and weekend.
Measuring accuracy threshold value described in the step (3) adopts square error to weigh, and square error is defined as follows:
Y wherein
iBe the actual value of arrival time,
Be the predicted value of SVM model, N is the quantity of SVM model measurement sample.
Filter state equation and observation equation described in the step (5) are respectively x
K+1=Fx
k+ u
k+ w
kAnd y
k=Hx
k+ v
k, wherein, x
k=[t
ks
k]
T,
H=[0 1],
t
kThe expression site k arrives the actual run time of prediction purpose website, T
K, k+1Actual run time between expression site k and the k+1, s
kRepresentative from the Source Site to the actual run time of website k, y
kExpression from the Source Site to the observation working time of website k, { w
k, v
kRepresent the system noise of statistical property the unknown respectively and measure noise.
The present invention is based on SVM and H
∞The advantage and the good effect of the dynamic bus arrival time forecasting methods of filtering are:
1 the present invention is divided into static prediction and two parts of performance prediction, taken all factors into consideration historical data and real-time information influence to the arrival time prediction, overcome based on the historical data method and can't reflect dynamic traffic state variation and the shortcoming that limited by external environment and hardware condition based on the real time data method, improved the accuracy of bus arrival time prediction.
2 the present invention utilize H first
∞The thought of filtering is carried out the bus arrival time prediction, and noise statistics is not made any hypothesis, makes its more realistic traffic stream characteristics complicated and changeable, has improved the robustness of bus arrival time prediction.
3 the present invention adopt offline mode with SVM model construction and test, and it is long to have overcome the online structure time of support vector machine, can't be applied to the shortcoming of actual prediction, has improved the ageing of bus arrival time prediction.
The invention has the beneficial effects as follows that this method has improved accuracy, robustness and the actual effect of bus arrival time prediction, for public transit system informatization and intelligent transportation development provides useful help.
Description of drawings:
Fig. 1 is a Forecasting Methodology process flow diagram of the present invention.
Fig. 2 is SVM model structure figure of the present invention.
Wherein, 201, input layer, realize the variable input of SVM model construction, 202, core layer, realize the structure and the test of SVM model, 203, output layer, the working time between the output public bus network adjacent sites.
Concrete enforcement:
Below in conjunction with accompanying drawing the embodiment of the invention is further described, present embodiment is illustrative, rather than determinate, can not limit protection scope of the present invention according to following examples.
A kind of based on SVM and H
∞The dynamic bus arrival time forecasting methods of filtering as depicted in figs. 1 and 2, may further comprise the steps:
(1) arrives the historical time database of each website according to bus, raw data is carried out pre-service, preprocessed data is put in order the data set that obtains meeting the preset recording form;
(2) by the method for statistical study the data set that step (1) obtains is classified, obtain flat peak class data set on working day, working day peak class data set, weekend flat peak class data set and weekend peak class data set;
(3) choose the input variable of SVM model construction, adopt the SVM method that the data set that classification in the step (2) obtains is carried out SVM model construction and test, set the measuring accuracy threshold value of SVM model, if the measuring accuracy of SVM model reaches the threshold value requirement, the SVM model construction is finished, otherwise rebuilds the SVM model; Described input variable comprises weather, highway section, time period and date;
Behind the historical time database update, repeating step (1), (2), (3) rebuild and test the SVM model;
(4) determine the numerical value of input variable, make up the SVM model of finishing according to step (3), obtain on the public bus network between the adjacent sites working time;
(5) according to filter state equation and observation equation, the real time execution information that automatic station name announcing system provides on working time on the public bus network that integrating step (4) obtains between the adjacent sites and the bus adopts H
∞Filtering method obtains the time that bus arrives each website of downstream, realizes dynamic bus arrival time prediction; Described real time execution information comprises that vehicle reaches website time, public transport operation line number, public transport arrival station period.
The pre-service of the raw data described in the step (1) comprises two stages: the processing stage of obliterated data and the journey time calculation stages; Be according to same vehicle, two record fields of same line raw data to be sorted the processing stage of obliterated data, whether continuous by detecting the website number field, whether judgment data loses; If lose the record strip number greater than 3, comprise 3, then abandon all station datas that comprise this record; If losing the record strip number is less than 3, then pass through the length computation polishing in adjacent upstream website and downstream website average running speed and current highway section; The journey time calculation stages is to utilize adjacent two website arrival times to subtract each other to obtain.
Data record format described in the step (1) comprises a plurality of fields, as Link Travel Time, weather, highway section, flat peak time section, rush hour section, working day and weekend.
Measuring accuracy threshold value described in the step (3) adopts square error to weigh, and square error is defined as follows:
Y wherein
iBe the actual value of arrival time,
Be the predicted value of SVM model, N is the quantity of SVM model measurement sample.
Filter state equation and observation equation described in the step (5) are respectively x
K+1=Fx
k+ u
k+ w
kAnd y
k=Hx
k+ v
k, wherein, x
k=[t
ks
k]
T,
H=[0 1],
t
kThe expression site k arrives the actual run time of prediction purpose website, T
K, k+1Actual run time between expression site k and the k+1, s
kRepresentative from the Source Site to the actual run time of website k, y
kExpression from the Source Site to the observation working time of website k, { w
k, v
kRepresent the system noise of statistical property the unknown respectively and measure noise.
Above-mentioned part is preferred embodiment of the present invention, is not to be used for limiting practical range of the present invention.Every equivalence of doing according to content of the present invention changes and modifies, and all should belong to content of the present invention.
Claims (5)
1. one kind based on SVM and H
∞The dynamic bus arrival time forecasting methods of filtering is characterized in that Forecasting Methodology may further comprise the steps:
(1) arrives the historical time database of each website according to bus, raw data is carried out pre-service, preprocessed data is put in order the data set that obtains meeting the preset recording form;
(2) by the method for statistical study the data set that step (1) obtains is classified, obtain flat peak class data set on working day, working day peak class data set, weekend flat peak class data set and weekend peak class data set;
(3) choose the input variable of SVM model construction, adopt the SVM method that the data set that classification in the step (2) obtains is carried out SVM model construction and test, set the measuring accuracy threshold value of SVM model, if the measuring accuracy of SVM model reaches the threshold value requirement, the SVM model construction is finished, otherwise rebuilds the SVM model; Described input variable comprises weather, highway section, time period and date;
Behind the historical time database update, repeating step (1), (2), (3) rebuild and test the SVM model;
(4) determine the numerical value of input variable, make up the SVM model of finishing according to step (3), obtain on the public bus network between the adjacent sites working time;
(5) according to filter state equation and observation equation, the real time execution information that automatic station name announcing system provides on working time on the public bus network that integrating step (4) obtains between the adjacent sites and the bus adopts H
∞Filtering method obtains the time that bus arrives each website of downstream, realizes dynamic bus arrival time prediction; Described real time execution information comprises that vehicle reaches website time, public transport operation line number, public transport arrival station period.
2. according to claim 1 a kind of based on SVM and H
∞The dynamic bus arrival time forecasting methods of filtering is characterized in that, the pre-service of the raw data described in the step (1) comprises two stages: the processing stage of obliterated data and the journey time calculation stages; Be according to same vehicle, two record fields of same line raw data to be sorted the processing stage of obliterated data, whether continuous by detecting the website number field, whether judgment data loses; If lose the record strip number greater than 3, comprise 3, then abandon all station datas that comprise this record; If losing the record strip number is less than 3, then pass through the length computation polishing in adjacent upstream website and downstream website average running speed and current highway section; The journey time calculation stages is to utilize adjacent two website arrival times to subtract each other to obtain.
3. according to claim 1 a kind of based on SVM and H
∞The dynamic bus arrival time forecasting methods of filtering is characterized in that the data record format described in the step (1) comprises a plurality of fields, as Link Travel Time, weather, highway section, flat peak time section, rush hour section, working day and weekend.
4. according to claim 1 a kind of based on SVM and H
∞The dynamic bus arrival time forecasting methods of filtering is characterized in that, the measuring accuracy threshold value described in the step (3) adopts square error to weigh, and square error is defined as follows:
Y wherein
iBe the actual value of arrival time,
Be the predicted value of SVM model, N is the quantity of SVM model measurement sample.
5. according to claim 1 a kind of based on SVM and H
∞The dynamic bus arrival time forecasting methods of filtering is characterized in that filter state equation and observation equation described in the step (5) are respectively x
K+1=Fx
k+ u
k+ w
kAnd y
k=Hx
k+ v
k, wherein, x
k=[t
ks
k]
T,
H=[0 1],
t
kThe expression site k arrives the actual run time of prediction purpose website, T
K, k+1Actual run time between expression site k and the k+1, s
kRepresentative from the Source Site to the actual run time of website k, y
kExpression from the Source Site to the observation working time of website k, { w
k, v
kRepresent the system noise of statistical property the unknown respectively and measure noise.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 201110028817 CN102074124B (en) | 2011-01-27 | 2011-01-27 | Dynamic bus arrival time prediction method based on support vector machine (SVM) and H-infinity filtering |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 201110028817 CN102074124B (en) | 2011-01-27 | 2011-01-27 | Dynamic bus arrival time prediction method based on support vector machine (SVM) and H-infinity filtering |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN102074124A true CN102074124A (en) | 2011-05-25 |
| CN102074124B CN102074124B (en) | 2013-05-08 |
Family
ID=44032648
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN 201110028817 Active CN102074124B (en) | 2011-01-27 | 2011-01-27 | Dynamic bus arrival time prediction method based on support vector machine (SVM) and H-infinity filtering |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102074124B (en) |
Cited By (25)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102708701A (en) * | 2012-05-18 | 2012-10-03 | 中国科学院信息工程研究所 | System and method for predicting arrival time of buses in real time |
| CN103310651A (en) * | 2013-05-24 | 2013-09-18 | 北京市交通信息中心 | Bus arrival prediction method based on real-time traffic status information |
| CN103578267A (en) * | 2012-07-18 | 2014-02-12 | 北京掌城科技有限公司 | Bus arrival predication method and system based on bus GPS data |
| CN103678917A (en) * | 2013-12-13 | 2014-03-26 | 杭州易和网络有限公司 | Bus real-time arrival time predicting method based on simulated annealing algorithm |
| WO2014146812A1 (en) * | 2013-03-22 | 2014-09-25 | Here Global B.V. | Vehicle arrival prediction |
| CN104217605A (en) * | 2013-05-31 | 2014-12-17 | 张伟伟 | Bus arrival time estimation method and device |
| CN104318757A (en) * | 2014-11-03 | 2015-01-28 | 大连海事大学 | Operation time forecasting method of buses on road segments of bus lanes |
| CN104424812A (en) * | 2013-08-28 | 2015-03-18 | 中国电信股份有限公司 | Bus arrival time prediction system and method |
| CN104781863A (en) * | 2012-11-15 | 2015-07-15 | 三菱电机株式会社 | Method for predicting future travel time on link |
| CN105303246A (en) * | 2015-09-07 | 2016-02-03 | 天津市市政工程设计研究院 | Multiline arrival time prediction for public transportation |
| CN105632222A (en) * | 2015-10-14 | 2016-06-01 | 上海博协软件有限公司 | Method and system for predicting station arrival time |
| CN106274994A (en) * | 2016-08-30 | 2017-01-04 | 湖南中车时代通信信号有限公司 | A kind of train arrival time Forecasting Methodology and system completely |
| CN107092988A (en) * | 2017-04-21 | 2017-08-25 | 北方工业大学 | Method for predicting station-parking time of bus on special lane |
| CN107784836A (en) * | 2016-08-25 | 2018-03-09 | 罗伯特·博世有限公司 | Method and apparatus for providing predicted vehicle movement |
| CN108230722A (en) * | 2017-12-02 | 2018-06-29 | 山东大学 | The accurate space-time bus platform instant messaging services fusion treatment method of work of the Big Dipper and system and device |
| CN108376477A (en) * | 2017-05-02 | 2018-08-07 | 山东大学 | Bus or train route cooperates with public transport precision data fusion center system and working method |
| CN109064742A (en) * | 2018-08-03 | 2018-12-21 | 北京航空航天大学 | A kind of adaptive public transport arrival time prediction technique based on SVM |
| CN109191845A (en) * | 2018-09-28 | 2019-01-11 | 吉林大学 | A kind of public transit vehicle arrival time prediction technique |
| CN109409598A (en) * | 2018-10-23 | 2019-03-01 | 河南工业大学 | Link travel time prediction method and device based on SVM and Kalman filtering |
| CN109544908A (en) * | 2018-10-24 | 2019-03-29 | 佛山市慧城信息科技有限公司 | A kind of method, electronic equipment and the storage medium of real-time prediction public transport arrival time |
| CN109767640A (en) * | 2017-11-09 | 2019-05-17 | 深圳大学 | One kind being based on the matched public transport operation time forecasting methods of historical data |
| CN110211379A (en) * | 2019-05-27 | 2019-09-06 | 南京航空航天大学 | A kind of public transport method for optimizing scheduling based on machine learning |
| CN111523560A (en) * | 2020-03-18 | 2020-08-11 | 第四范式（北京）技术有限公司 | Training method, prediction method, device and system for number prediction model of arriving trucks |
| CN111667689A (en) * | 2020-05-06 | 2020-09-15 | 浙江师范大学 | Method, device and computer device for predicting vehicle travel time |
| CN115662176A (en) * | 2022-12-13 | 2023-01-31 | 天津市政工程设计研究总院有限公司 | Flexible bus scheduling optimization method based on robust optimization |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109584552B (en) * | 2018-11-28 | 2021-04-30 | 青岛大学 | Bus arrival time prediction method based on network vector autoregressive model |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101383090A (en) * | 2008-10-24 | 2009-03-11 | 北京航空航天大学 | Floating vehicle information processing method under parallel road network structure |
| KR20090126545A (en) * | 2008-06-04 | 2009-12-09 | 주식회사 만도 | Pedestrian recognition method and apparatus using a posteriori probability and pedestrian protection apparatus using same |
| CN101615340A (en) * | 2009-07-24 | 2009-12-30 | 北京工业大学 | Real-time information processing method in the bus dynamic dispatching |
-
2011
- 2011-01-27 CN CN 201110028817 patent/CN102074124B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20090126545A (en) * | 2008-06-04 | 2009-12-09 | 주식회사 만도 | Pedestrian recognition method and apparatus using a posteriori probability and pedestrian protection apparatus using same |
| CN101383090A (en) * | 2008-10-24 | 2009-03-11 | 北京航空航天大学 | Floating vehicle information processing method under parallel road network structure |
| CN101615340A (en) * | 2009-07-24 | 2009-12-30 | 北京工业大学 | Real-time information processing method in the bus dynamic dispatching |
Non-Patent Citations (1)
| Title |
|---|
| 《中国公路学报》 20080315 于滨等 "基于SVM和Kalman滤波的公交车到站时间预测模型" 89-97 第21卷, 第2期 * |
Cited By (34)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102708701B (en) * | 2012-05-18 | 2015-01-28 | 中国科学院信息工程研究所 | System and method for predicting arrival time of buses in real time |
| CN102708701A (en) * | 2012-05-18 | 2012-10-03 | 中国科学院信息工程研究所 | System and method for predicting arrival time of buses in real time |
| CN103578267A (en) * | 2012-07-18 | 2014-02-12 | 北京掌城科技有限公司 | Bus arrival predication method and system based on bus GPS data |
| CN104781863A (en) * | 2012-11-15 | 2015-07-15 | 三菱电机株式会社 | Method for predicting future travel time on link |
| WO2014146812A1 (en) * | 2013-03-22 | 2014-09-25 | Here Global B.V. | Vehicle arrival prediction |
| US9020754B2 (en) | 2013-03-22 | 2015-04-28 | Here Global B.V. | Vehicle arrival prediction |
| CN103310651B (en) * | 2013-05-24 | 2016-03-02 | 北京市交通信息中心 | A kind of public transport based on real-time road condition information is arrived at a station Forecasting Methodology |
| CN103310651A (en) * | 2013-05-24 | 2013-09-18 | 北京市交通信息中心 | Bus arrival prediction method based on real-time traffic status information |
| CN104217605A (en) * | 2013-05-31 | 2014-12-17 | 张伟伟 | Bus arrival time estimation method and device |
| CN104424812A (en) * | 2013-08-28 | 2015-03-18 | 中国电信股份有限公司 | Bus arrival time prediction system and method |
| CN103678917A (en) * | 2013-12-13 | 2014-03-26 | 杭州易和网络有限公司 | Bus real-time arrival time predicting method based on simulated annealing algorithm |
| CN103678917B (en) * | 2013-12-13 | 2016-11-23 | 杭州易和网络有限公司 | A kind of real-time arrival time Forecasting Methodology of public transport based on simulated annealing |
| CN104318757A (en) * | 2014-11-03 | 2015-01-28 | 大连海事大学 | Operation time forecasting method of buses on road segments of bus lanes |
| CN105303246A (en) * | 2015-09-07 | 2016-02-03 | 天津市市政工程设计研究院 | Multiline arrival time prediction for public transportation |
| CN105632222B (en) * | 2015-10-14 | 2018-04-17 | 上海博协软件有限公司 | Forecast the method and its system of arrival time |
| CN105632222A (en) * | 2015-10-14 | 2016-06-01 | 上海博协软件有限公司 | Method and system for predicting station arrival time |
| CN107784836A (en) * | 2016-08-25 | 2018-03-09 | 罗伯特·博世有限公司 | Method and apparatus for providing predicted vehicle movement |
| CN106274994A (en) * | 2016-08-30 | 2017-01-04 | 湖南中车时代通信信号有限公司 | A kind of train arrival time Forecasting Methodology and system completely |
| CN107092988A (en) * | 2017-04-21 | 2017-08-25 | 北方工业大学 | Method for predicting station-parking time of bus on special lane |
| CN107092988B (en) * | 2017-04-21 | 2020-04-10 | 北方工业大学 | Method for predicting station-parking time of bus on special lane |
| CN108376477A (en) * | 2017-05-02 | 2018-08-07 | 山东大学 | Bus or train route cooperates with public transport precision data fusion center system and working method |
| CN109767640A (en) * | 2017-11-09 | 2019-05-17 | 深圳大学 | One kind being based on the matched public transport operation time forecasting methods of historical data |
| CN108230722A (en) * | 2017-12-02 | 2018-06-29 | 山东大学 | The accurate space-time bus platform instant messaging services fusion treatment method of work of the Big Dipper and system and device |
| CN109064742A (en) * | 2018-08-03 | 2018-12-21 | 北京航空航天大学 | A kind of adaptive public transport arrival time prediction technique based on SVM |
| CN109191845A (en) * | 2018-09-28 | 2019-01-11 | 吉林大学 | A kind of public transit vehicle arrival time prediction technique |
| CN109191845B (en) * | 2018-09-28 | 2020-09-25 | 吉林大学 | Bus arrival time prediction method |
| CN109409598A (en) * | 2018-10-23 | 2019-03-01 | 河南工业大学 | Link travel time prediction method and device based on SVM and Kalman filtering |
| CN109544908A (en) * | 2018-10-24 | 2019-03-29 | 佛山市慧城信息科技有限公司 | A kind of method, electronic equipment and the storage medium of real-time prediction public transport arrival time |
| CN109544908B (en) * | 2018-10-24 | 2022-04-15 | 佛山市慧城信息科技有限公司 | Method for predicting bus arrival time in real time, electronic equipment and storage medium |
| CN110211379A (en) * | 2019-05-27 | 2019-09-06 | 南京航空航天大学 | A kind of public transport method for optimizing scheduling based on machine learning |
| CN111523560A (en) * | 2020-03-18 | 2020-08-11 | 第四范式（北京）技术有限公司 | Training method, prediction method, device and system for number prediction model of arriving trucks |
| CN111667689A (en) * | 2020-05-06 | 2020-09-15 | 浙江师范大学 | Method, device and computer device for predicting vehicle travel time |
| CN115662176A (en) * | 2022-12-13 | 2023-01-31 | 天津市政工程设计研究总院有限公司 | Flexible bus scheduling optimization method based on robust optimization |
| CN115662176B (en) * | 2022-12-13 | 2023-05-26 | 天津市政工程设计研究总院有限公司 | Flexible bus dispatching optimization method based on robust optimization |
Also Published As
| Publication number | Publication date |
|---|---|
| CN102074124B (en) | 2013-05-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102074124B (en) | Dynamic bus arrival time prediction method based on support vector machine (SVM) and H-infinity filtering | |
| CN114091359B (en) | Reservoir water and sand prediction model training and reservoir water and sand prediction method and device | |
| CN113487066B (en) | Long-time-sequence freight volume prediction method based on multi-attribute enhanced graph convolution-Informer model | |
| CN111104981A (en) | Hydrological prediction precision evaluation method and system based on machine learning | |
| CN102800198B (en) | Measuring and calculating method for traffic flow of section of expressway | |
| CN102110365A (en) | Road condition prediction method and road condition prediction system based on space-time relationship | |
| CN105513337A (en) | Passenger flow volume prediction method and device | |
| CN103413443A (en) | Short-term traffic flow forecasting method based on hidden Markov model | |
| Huang et al. | A new air quality forecasting model using data mining and artificial neural network | |
| CN104320789A (en) | Game theory-based vehicle networking RSU optimal configuration method | |
| CN103456163A (en) | City expressway interchange traffic capacity and running status discrimination method and system | |
| CN112419711B (en) | Closed parking lot parking demand prediction method based on improved GMDH algorithm | |
| Du et al. | Traffic events oriented dynamic traffic assignment model for expressway network: a network flow approach | |
| Wang et al. | Has the healthy city pilot policy improved urban health development performance in China? Evidence from a quasi-natural experiment | |
| CN111199298A (en) | Flood forecasting method and system based on neural network | |
| Shan et al. | Probabilistic model for vehicle trajectories reconstruction using sparse mobile sensor data on freeways | |
| Xia et al. | Deeprailway: a deep learning system for forecasting railway traffic | |
| CN106203887A (en) | A kind of network of highways characteristic analysis method based on cross classification and device | |
| CN109447352B (en) | Bus trip OD matrix reverse-thrust combination method | |
| Liu et al. | A hybrid model based on kalman filter and neutral network for traffic prediction | |
| Guo et al. | Traffic flow fluctuation analysis based on beijing taxi gps data | |
| Engelson et al. | The role of volume-delay functions in forecasting and evaluating congestion charging schemes: the Stockholm case | |
| Zhang | Review on Traffic Flow Prediction Method Based on Neural Network | |
| CN101697250A (en) | Modified method for predicting section, travel route and time of city bus | |
| Fangming et al. | Simulation of traffic incident detection based on VISSIM and neural network |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant |