CN103500362A - Urban road speed prediction method based on spectral analysis - Google Patents
Urban road speed prediction method based on spectral analysis Download PDFInfo
- Publication number
- CN103500362A CN103500362A CN201310390819.9A CN201310390819A CN103500362A CN 103500362 A CN103500362 A CN 103500362A CN 201310390819 A CN201310390819 A CN 201310390819A CN 103500362 A CN103500362 A CN 103500362A
- Authority
- CN
- China
- Prior art keywords
- matrix
- data
- interval
- days
- road
- 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
- 238000000034 method Methods 0.000 title claims abstract description 27
- 238000010183 spectrum analysis Methods 0.000 title abstract description 4
- 239000011159 matrix material Substances 0.000 claims description 51
- 238000001228 spectrum Methods 0.000 claims description 11
- 230000015572 biosynthetic process Effects 0.000 claims description 6
- 238000006467 substitution reaction Methods 0.000 claims description 6
- 238000000354 decomposition reaction Methods 0.000 claims description 5
- 241001269238 Data Species 0.000 claims description 3
- 230000003203 everyday effect Effects 0.000 claims description 3
- 238000006243 chemical reaction Methods 0.000 claims description 2
- 239000000203 mixture Substances 0.000 claims description 2
- YHXISWVBGDMDLQ-UHFFFAOYSA-N moclobemide Chemical compound C1=CC(Cl)=CC=C1C(=O)NCCN1CCOCC1 YHXISWVBGDMDLQ-UHFFFAOYSA-N 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 2
- 241001123248 Arma Species 0.000 description 1
- 230000008878 coupling Effects 0.000 description 1
- 238000010168 coupling process Methods 0.000 description 1
- 238000005859 coupling reaction Methods 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000004141 dimensional analysis Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000012067 mathematical method Methods 0.000 description 1
- 238000013178 mathematical model Methods 0.000 description 1
- 239000004576 sand Substances 0.000 description 1
- 238000000714 time series forecasting Methods 0.000 description 1
Images
Landscapes
- Traffic Control Systems (AREA)
Abstract
The invention relates to an urban road speed prediction method based on spectral analysis. The urban road speed prediction method includes the first step of collecting vehicle speed data collected by a GPS, the second step of selecting road speed model parameters based on the spectral analysis by using given data, and the third step of predicting unknown road speed according to a model formula. The urban road speed prediction method has the advantages that historical data needed by a training model are relatively less, the operation is relatively easy and convenient, and the parameters in the model can be adjusted rapidly and dynamically along with the change of road segments and time frames; the method can be suitable for the road segments with large traffic flow fluctuation and can be well matched with an actual road speed condition, and prediction accuracy and prediction reliability are improved.
Description
Technical field
The present invention relates to a kind of urban road speed predicting method based on analysis of spectrum, belong to the intelligent transportation system research category.
Background technology
The urban highway traffic Forecasting Methodology is by the road speeds of respective stretch in the prediction short time range, obtain corresponding road traffic state, and utilize the multiple channel issue to predict the outcome, induce driver and crew to select reasonable traffic path, can play the effect of alleviating traffic congestion, more and more receive publicity in recent years.
Analysis of spectrum is spatial data to be carried out to a kind of mathematical method of general layout, dimensional analysis, can present powerful expressive ability, even a single point on road, also can effectively show its real traffic, thereby the impact that Reduce variation brings, be applicable to analyze the time series that fluctuation is larger.But directly utilize the method for spectral analysis technology predicted city road speeds have not been reported.
At present, the main method of road speeds prediction comprises ARIMA(difference ARMA model), Kalman filter model etc.ARIMA is a kind of fully Time Series Forecasting Methods based on empirical statistics, and it is passed forecasting object in time and the random series that forms is carried out approximate description with certain mathematical model, and utilizes the past value of this sequence and present value to carry out the predict future value.Kalman filter model adopts the optimum criterion of least mean-square error as prediction, utilizes the predicted value of previous moment and the observed reading of current time to upgrade the prediction to state variable, obtains the predicted value of current time.Although these two kinds of models can comparatively fast be predicted unknown data, are not suitable for the time series that fluctuation is larger.In city road network, road traffic state exists and changes the characteristics such as fast, that fluctuation range is large, and said method can't reflect these variations in time, causes predicated error larger.
Summary of the invention
The present invention will overcome the deficiencies in the prior art, proposes a kind of urban road speed predicting method based on analysis of spectrum, alleviates the impact of urban road traffic state rapid fluctuations on the road speeds prediction, improves forecasting accuracy.
For achieving the above object, at first the present invention collects road historical data and prediction given data on the same day, recycles both corresponding interval relations and selects prediction model parameters, and last bind profile analytical model prediction obtains unknown data.
Method of the present invention realizes by following steps:
Step 1. is collected the vehicle speed data that GPS collects
GPS collects data such as comprising acquisition time, car number, car speed, wherein chooses vehicle speed data as source data;
Step 2. preference pattern parameter
Mean the relation of data to be predicted historical data corresponding with it by given data and the relation of its corresponding historical data on prediction same day, refer to the parameter matrix C in forecast model;
Wherein the road speeds forecast model formula based on analysis of spectrum is as follows:
In formula:
mean certain continuum road speeds matrix that prediction obtains; C representation parameter matrix; φ is the eigenvectors matrix of describing respective bins road speeds variation tendency;
Ask parameter matrix C concrete steps as follows:
2.1 at first the GPS car speed is converted into to road speeds by map-matching method;
2.2 then on average be divided into to N interval every day, known historical continuous N sky N interval and road speeds data j interval before M+1 days;
2.3 define the matrix that historical continuous N sky N interval data are the M*N rank, be denoted as X.According to R=X
tx tries to achieve the covariance matrix R (N*N) of matrix X;
2.4 adopt QR decomposition method (matrix decomposition becomes an orthonomal matrix Q and upper triangular matrix R) to try to achieve the eigenvectors matrix of covariance matrix R, be designated as φ (N*N).Choose N*K and partly be designated as φ ' in eigenvectors matrix φ, wherein, K (1≤K<N) means exponent number, the quantity of the proper vector of using.Choose in matrix φ ' the 1st and walk to the capable part of j, be designated as φ
1(j*K); In like manner, in φ ', j+1 walks to the capable part of N, is designated as φ
2((N-j) * K);
2.5 the matrix of M+1 days known interval censored data compositions is denoted as
the φ tried to achieve with previous step
1the formula that together substitution model formation (1) conversion obtains
in, can try to achieve the parameter matrix meaned between (interval 1 to interval j) data between M+1 days known zone and corresponding M days historical interval censored datas, be designated as C
1(1*K), also can be used as the parameter matrix of (after interval j, not comprising interval j) data and historical M days interval censored data between M+1 days unknown areas;
Step 3. is according to model formation predicted link speed
By parameter matrix C obtained in the previous step
1and φ
2in while substitution formula (1), the measurable M+1 days later road speeds of interval j.
Remarkable result of the present invention is: the required historical data of urban road speed prediction model based on analysis of spectrum is less, model parameter can be made dynamic adjustment fast with the variation of highway section, period, better with real road speed condition coupling, and can alleviate the impact of urban road traffic state rapid fluctuations on the road speeds prediction, improve forecasting accuracy.
The accompanying drawing explanation
Fig. 1 is based on the urban road speed predicting method process flow diagram of analysis of spectrum.
Embodiment
Below in conjunction with drawings and Examples, technical scheme of the present invention is described in further detail.Following examples are implemented take technical scheme of the present invention under prerequisite, provided detailed embodiment and process, but protection scope of the present invention are not limited to following embodiment.
The concrete implementation step of the present embodiment is as follows:
Step 1. utilizes GPS to collect the road speeds data
1.1GPS collect data such as comprising acquisition time, car number, car speed, wherein choose vehicle speed data as source data.
1.2 using Lu— west part of the ring road, city of Hangzhou stadium, west part of the ring road-,Bao Chu road, white sand road-40 highway sections such as —Feng Qi road, Bei Shanlu ,Bao Chu road as the data acquisition zone, 00:00-23:59 is as the data acquisition time section, and the vehicle speed data that GPS collected every 15 minutes is as source data.
1.3 it is internal memory 2G that data are stored in hardware environment, on the PC of hard disk 300G.
Step 2. preference pattern parameter
Mean the relation of data to be predicted historical data corresponding with it by given data and the relation of its corresponding historical data on prediction same day, refer to the parameter matrix C in forecast model.
Wherein the road speeds forecast model formula based on analysis of spectrum is as follows:
In formula:
mean certain continuum road speeds matrix that prediction obtains; C representation parameter matrix; φ is the eigenvectors matrix of describing respective bins road speeds variation tendency.
Ask parameter matrix C concrete steps as follows:
2.1 at first the GPS car speed is converted into to road speeds by map-matching method.
2.2 then on average be divided into to 96 intervals (every 15 minutes being an interval) every day, 192 intervals in this example on September 10th, 1 and continuous 2 days of September 11 and September the 1st to the 3rd interval on the 12nd the road speeds data.Choose 6 interval censored data declarative procedures that comprise in initial 00:00-01:30, wherein exponent number K gets 2.
2.3 historical continuous 2 days 6 interval censored datas of definition form matrix X ' (2*6).
According to formula R=X
tx tries to achieve covariance matrix R (6*6), as follows:
2.4 adopt the QR decomposition method to try to achieve the eigenvectors matrix φ (6*6) of R, as follows:
2.5 selected part eigenvectors matrix φ ' is (6*2), as follows from the eigenvectors matrix φ of corresponding window:
2.6 choose respectively the eigenvectors matrix φ that means the road speeds variation tendency in front 3 intervals and rear 3 intervals in this matrix from φ '
1(3*2) and φ
2(3*2).As follows respectively:
2.7 be below the matrixes that in the 3rd day, the data in front 3 the corresponding intervals of window form therewith
{50?42?41}
Will
and φ
1substitution formula (1), calculate C (1*2), and result is as follows:
{-89.52?14.66}
Step 3. is according to model formation predicted link speed
By φ
2in the parameter matrix C substitution model formation (1) of trying to achieve in step b, i.e. data of rear 3 the unknowns that window is corresponding therewith on measurable September 12, result is as follows:
{21?35?25}
The present invention, on 40 highway sections of Hangzhou major trunk roads, has gathered the gps data checking of 30 days.Experimental result shows, with adopting ARIMA, with kalman filter method predicted link speed, compares, and the Forecasting Methodology based on analysis of spectrum makes predicated error (RMSE: root-mean-square error) reduce more than 40%.
Claims (1)
1. the urban road speed predicting method based on analysis of spectrum is characterized in that comprising the following steps:
Step 1. is collected the vehicle speed data that GPS collects
GPS collects data such as comprising acquisition time, car number, car speed, wherein chooses vehicle speed data as source data;
Step 2. preference pattern parameter
Mean the relation of data to be predicted historical data corresponding with it by given data and the relation of its corresponding historical interval censored data on prediction same day, refer to the parameter matrix C in forecast model;
Wherein the road speeds forecast model formula based on analysis of spectrum is as follows:
In formula
mean certain continuum road speeds matrix that prediction obtains; C representation parameter matrix; φ is the eigenvectors matrix of describing respective bins road speeds variation tendency;
Ask parameter matrix C concrete steps as follows:
2.1 at first the GPS car speed is converted into to road speeds by map-matching method;
2.2 then on average be divided into to N interval every day, known historical continuous N sky N interval and road speeds data j interval before M+1 days;
2.3 define the matrix that historical continuous N sky N interval data are the M*N rank, be denoted as X.According to R=X
tx tries to achieve the covariance matrix R (N*N) of matrix X;
Try to achieve the eigenvectors matrix of covariance matrix R 2.4 adopt QR decomposition method (matrix decomposition becomes an orthonomal matrix Q and upper triangular matrix R), be designated as φ (N*N), choose N*K and partly be designated as φ ' in eigenvectors matrix φ, wherein, K (1≤K<N) means exponent number, the quantity of the proper vector of using; Choose in matrix φ ' the 1st and walk to the capable part of j, be designated as φ
1(j*K); In like manner, in φ ', j+1 walks to the capable part of N, is designated as φ
2((N-j) * K);
2.5 the matrix of M+1 days known interval censored data compositions is denoted as
the φ tried to achieve with previous step
1the formula that together substitution model formation (1) conversion obtains
in, can try to achieve the parameter matrix meaned between (interval 1 to interval j) data between M+1 days known zone and corresponding M days historical interval censored datas, be designated as C
1(1*K), also can be used as the parameter matrix of (after interval j, not comprising interval j) data and historical M days interval censored data between M+1 days unknown areas;
Step 3. is according to model formation predicted link speed
By parameter matrix C obtained in the previous step
1and φ
2in while substitution formula (1), the measurable M+1 days later road speeds of interval j.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201310390819.9A CN103500362B (en) | 2013-08-30 | 2013-08-30 | A kind of urban road speed predicting method based on analysis of spectrum |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201310390819.9A CN103500362B (en) | 2013-08-30 | 2013-08-30 | A kind of urban road speed predicting method based on analysis of spectrum |
Publications (2)
Publication Number | Publication Date |
---|---|
CN103500362A true CN103500362A (en) | 2014-01-08 |
CN103500362B CN103500362B (en) | 2016-08-17 |
Family
ID=49865568
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201310390819.9A Expired - Fee Related CN103500362B (en) | 2013-08-30 | 2013-08-30 | A kind of urban road speed predicting method based on analysis of spectrum |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN103500362B (en) |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107993452A (en) * | 2017-12-20 | 2018-05-04 | 夏莹杰 | Speed-measuring method based on road passage rate on WIFI probes detection highway |
CN105405127B (en) * | 2015-10-30 | 2018-06-01 | 长安大学 | A kind of highway minibus speed of service Forecasting Methodology |
CN109711440A (en) * | 2018-12-13 | 2019-05-03 | 新奥数能科技有限公司 | A kind of data exception detection method and device |
CN111653084A (en) * | 2019-07-26 | 2020-09-11 | 银江股份有限公司 | Short-term traffic flow prediction method based on space-time feature selection and Kalman filtering |
CN112950926A (en) * | 2019-12-10 | 2021-06-11 | 宁波中国科学院信息技术应用研究院 | Urban trunk road speed prediction method based on big data and deep learning |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2006285493A (en) * | 2005-03-31 | 2006-10-19 | Daihatsu Motor Co Ltd | Device and method for estimating road model |
CN102610092A (en) * | 2012-03-23 | 2012-07-25 | 天津大学 | Urban road speed predication method based on RBF (radial basis function) neural network |
-
2013
- 2013-08-30 CN CN201310390819.9A patent/CN103500362B/en not_active Expired - Fee Related
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2006285493A (en) * | 2005-03-31 | 2006-10-19 | Daihatsu Motor Co Ltd | Device and method for estimating road model |
CN102610092A (en) * | 2012-03-23 | 2012-07-25 | 天津大学 | Urban road speed predication method based on RBF (radial basis function) neural network |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105405127B (en) * | 2015-10-30 | 2018-06-01 | 长安大学 | A kind of highway minibus speed of service Forecasting Methodology |
CN107993452A (en) * | 2017-12-20 | 2018-05-04 | 夏莹杰 | Speed-measuring method based on road passage rate on WIFI probes detection highway |
CN109711440A (en) * | 2018-12-13 | 2019-05-03 | 新奥数能科技有限公司 | A kind of data exception detection method and device |
CN111653084A (en) * | 2019-07-26 | 2020-09-11 | 银江股份有限公司 | Short-term traffic flow prediction method based on space-time feature selection and Kalman filtering |
CN112950926A (en) * | 2019-12-10 | 2021-06-11 | 宁波中国科学院信息技术应用研究院 | Urban trunk road speed prediction method based on big data and deep learning |
Also Published As
Publication number | Publication date |
---|---|
CN103500362B (en) | 2016-08-17 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN105513350A (en) | Time-phased multi-parameter short-term traffic flow prediction method based on time-space characteristics | |
CN101794507B (en) | Method for evaluating macroscopic road network traffic state based on floating car data | |
CN104157139B (en) | A kind of traffic congestion Forecasting Methodology and method for visualizing | |
CN102610092A (en) | Urban road speed predication method based on RBF (radial basis function) neural network | |
CN104217250B (en) | A kind of urban rail transit new line based on historical data opens passenger flow forecasting | |
CN104299442B (en) | A kind of city Route Travel Time Estimation Method based on pattern match | |
CN103500362A (en) | Urban road speed prediction method based on spectral analysis | |
CN109840660A (en) | A kind of vehicular characteristics data processing method and vehicle risk prediction model training method | |
CN106951999A (en) | The modeling of a kind of travel modal and the moment Combination selection that sets out and analysis method | |
CN104933862A (en) | Urban traffic jam intelligent combination prediction method based on track of floating vehicle | |
CN107330217A (en) | A kind of middle sight oil consumption Forecasting Methodology based on RBFNN | |
CN107284452B (en) | Merge the hybrid vehicle future operating mode forecasting system of intelligent communication information | |
CN110310474A (en) | A kind of vehicle flowrate prediction technique and device based on space-time residual error network | |
CN103295414A (en) | Bus arrival time forecasting method based on mass historical GPS (global position system) trajectory data | |
CN104376384A (en) | Typhoon day maximum daily load prediction system based on power big data analysis | |
CN104269059A (en) | City path travel time forecasting method based on multi-source data fusion | |
CN103745602B (en) | A kind of traffic flow forecasting method average based on sliding window | |
CN105389640A (en) | Method for predicting suburban railway passenger flow | |
CN106370198A (en) | Route selection method taking outgoing delayed reaction into account | |
CN104240520A (en) | GSO (glowworm swarm optimization) based highway traffic prediction method | |
CN109377761A (en) | Traffic factor network establishing method based on Markov-chain model | |
CN106846891B (en) | A kind of Public Parking berth multistep forecasting method decomposed based on sequence | |
Kumar et al. | Analytical approach to identify the optimum inputs for a bus travel time prediction method | |
CN104504245A (en) | Method of utilizing GPS trip survey data to identify trips and activities | |
CN104182633A (en) | Hierarchical traffic operation evaluation method |
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 | ||
CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20160817 |
|
CF01 | Termination of patent right due to non-payment of annual fee |