CN112802335B - Intelligent traffic management method based on Beidou navigation system - Google Patents
Intelligent traffic management method based on Beidou navigation system Download PDFInfo
- Publication number
- CN112802335B CN112802335B CN202011623294.5A CN202011623294A CN112802335B CN 112802335 B CN112802335 B CN 112802335B CN 202011623294 A CN202011623294 A CN 202011623294A CN 112802335 B CN112802335 B CN 112802335B
- Authority
- CN
- China
- Prior art keywords
- traffic flow
- prediction
- road
- traffic
- module
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
- G08G1/0125—Traffic data processing
- G08G1/0129—Traffic data processing for creating historical data or processing based on historical data
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
- G06N20/20—Ensemble learning
-
- 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
- G08G1/0137—Measuring and analyzing of parameters relative to traffic conditions for specific applications
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/07—Controlling traffic signals
- G08G1/08—Controlling traffic signals according to detected number or speed of vehicles
Abstract
The invention relates to the technical field of traffic management, and provides an intelligent traffic management method based on a Beidou navigation system, which comprises a step S0 that a control center is provided with a traffic flow prediction module and a Beidou navigation module, wherein the traffic flow prediction module predicts traffic flow, the Beidou navigation module obtains the road congestion condition at the current moment, and the traffic congestion condition and the Beidou navigation module establish one-to-one mapping; s1, when the two do not meet the mapping relationship, the control center sends instructions to the road detection units, wherein the road detection units are provided with Beidou satellite navigation positioning modules; s2, after receiving the instructions, the control modules control the road detection unit to take a picture to acquire road picture information; s3, the control module transmits the road photo information with position information acquired by the Beidou satellite navigation and positioning module back to the control center; and S4, the control center formulates a road management scheme for the road photo information. The invention solves the problem that the theoretical and actual differences are large when the traffic flow is applied to the actual traffic management.
Description
Technical Field
The invention relates to the technical field of traffic methods, in particular to an intelligent traffic management method based on a Beidou navigation system.
Background
Increased usage of automobiles is a major cause of urban traffic congestion. Due to the convenience of automobiles, the traffic flow in urban areas is increasing day by day, and the traffic flow of working, traveling and shopping is gushed into the city center from all sides every peak time. However, the large disadvantage of the automobile is that the space is wasted, but the number of the automobiles is increased continuously, so that the existing road cannot load the traffic flow with the large amount, and the situation of blockage is caused.
For road traffic flow management, the difficulty is high, a plurality of uncontrollable factors exist, the traffic flow sometimes does not accord with the road congestion situation, and the traffic flow becomes small on the contrary when the reason is congested, so that the problem that the theoretical actual difference is large when the predicted traffic flow is applied to actual traffic management exists.
Disclosure of Invention
The invention provides an intelligent traffic management method based on a Beidou navigation system, which solves the problems in the related art.
The technical scheme of the invention is as follows:
an intelligent traffic management method based on a Beidou navigation system comprises the following steps
S0, the control center is provided with a traffic flow prediction module and a Beidou navigation module, the traffic flow prediction module predicts traffic flow, the Beidou navigation module obtains the road congestion condition at the current moment, and the two modules establish one-to-one mapping;
s1, when the two detection units do not meet the mapping relationship, determining a plurality of road detection units which may be involved, and sending instructions to the road detection units by a control center, wherein the road detection units are provided with a Beidou satellite navigation positioning module, a communication module and a camera module which are all connected with a control module;
s2, after receiving the instructions, the control modules control the road detection unit to take a picture to obtain road picture information;
s3, the control module transmits the road photo information with position information acquired by the Beidou satellite navigation and positioning module back to the control center;
and S4, the control center formulates a road management scheme for the road photo information.
Further, the road detection unit is a mobile road detection unit or a fixed road detection unit, and the mobile road detection unit is an unmanned aerial vehicle or a road driving device.
Furthermore, in S0, the traffic flow prediction method predicts the crossroad at t + m moment according to the traffic condition of the crossroad at t moment and t-m momentOne of them flows downstream by a flow Q Prediction (t + m) Therefore, different levels of monitoring of traffic flow are performed, and the method specifically comprises the following steps:
s01, determining the independent variable parameters of the training model as
t time three upstream traffic flows q in the direction of the crossroad 1 、q 2 、q 3 The information, obtained by video surveillance,
the downstream traffic flow Q in the direction of the crossroad at the time t and the time t-m (t) 、Q (t-m) The information, obtained by video surveillance,
the downstream vehicle traffic influence level L in this direction at time t at the intersection,
one side of the crossroad downstream traffic flow Q at the moment when the dependent variable parameter is t + m Prediction (t + m) ；
S02, recording historical data of the independent variable parameters and the dependent variable parameters;
s03, predicting to obtain a BP predicted value Q of traffic flow BP of one side of the crossroad downstream at the moment t + m through a BP neural network bp prediction (t + m) Specifically, historical data of respective variable parameters and dependent variable parameters are input into a BP neural network to obtain a BP neural network traffic model;
at the time of t, the numerical values of the respective variables are input into a BP neural network traffic model to obtain a BP predicted value Q of the downstream traffic flow in the direction of the crossroad at the time of t + m bp prediction (t + m) ，
Further, in step S01, the downstream vehicle traffic influence level L in the direction of the intersection at time t is specifically according to the following formula:
and li is from the moment t-m to the moment t, the passing vehicles of each vehicle are classified, and the cloud vehicle grades corresponding to the license plate numbers are acquired by the camera.
Further, the vehicle classification li is according to the following formula:
li＝a*r+b*s+c*lio，(i＝1,2,3…，n)，
wherein r is the accumulated number of roadside illegal parking in the driving process of the vehicle, s is the accumulated number of crossing intersections when the vehicle is in red light in the driving process, lio is the vehicle classification of other vehicles under the owner name of the vehicle, a, b and c are respective weights,
wherein, r and s are acquired by camera video acquisition.
Further, in S04, Q Prediction (t + m) ＝Q bp prediction (t + m) 。
Further, between S03 and S04, also include
S030, predicting to obtain an RF predicted value Q of traffic flow of one side of the crossroad to the downstream at the t + m moment through a random forest model rf prediction (t + m) Specifically, historical data of respective variable parameters and dependent variable parameters are input into the random forest model to obtain a random forest traffic model;
at the time of t, the numerical values of the respective variables are input into the random deep forest model to obtain a downstream traffic flow RF predicted value Q in the direction of the crossroad at the time of t + m rf prediction (t + m) 。
Further, among them, in S04, Q Prediction (t + m) ＝K 1 *Q bp prediction (t + m) +K 2 *Q rf prediction (t + m) 。
Further, in S04, K 1 、K 2 Is given as initial condition K 1 +K 2 =1; and K is 1 、K 2 And calculating by an optimal weighted combination method.
The working principle and the beneficial effects of the invention are as follows:
1. the traffic jam condition is divided into four conditions of smooth running, slow running, jam and serious jam according to the running speed of the road, correspondingly, the detected traffic flow is also divided into four conditions to be in one-to-one correspondence, and when the two conditions are inconsistent, the abnormal traffic condition can be well found.
2. In the invention, when the traffic flow is inconsistent with the road congestion condition, the function of the road detection unit is played, so that more direct and effective real road data is obtained, and intelligent and automatic image analysis is carried out according to the traffic condition, so that the road abnormity is found by utilizing big data, the automatic image analysis in all time periods is well realized, and the calculated amount is greatly reduced.
3. The method for predicting the traffic flow based on the combination of the random forest model RF and the BP neural network overcomes respective defects of the two models, the optimal weighted combination model is well utilized to be combined through the two modes, the prediction precision is further improved, the effectiveness after combination is shown through comparison with the non-combined mode, and the accuracy and precision of the traffic prediction are further improved
4. In the invention, the traffic flow data is preprocessed, and the traffic data is denoised by wavelet analysis, so that the information interference in the acquired data can be reduced, and the deviation can be well eliminated.
5. In the invention, the monitoring level M of the crossroad in the direction can be determined according to the road congestion condition, and can be divided into two levels or three levels, so that the monitoring level with low requirement can be carried out in the uncongested time period, and the monitoring level with high requirement can be carried out in the congested time period, thereby reasonably utilizing the monitoring resource and improving the resource integration degree.
6. In the invention, the dependent variable parameters are improved better, and three upstream traffic flows q in the direction of the crossroad at the moment t are selected 1 、q 2 、q 3 Downstream traffic flow Q in the direction of the crossroad at the time t and the time t-m (t) 、Q (t-m) And the influence level L of the downstream vehicle traffic in the direction of the crossroad at the moment t is integrally used as a dependent variable, so that the prediction effect of the BP neural network prediction model is well improved.
Drawings
The invention is described in further detail below with reference to the drawings and the detailed description.
FIG. 1 is a comparison diagram of a predicted value and an actual value of a traffic flow according to an embodiment of the present invention when predicting a traffic flow;
FIG. 2 is a comparison graph of a predicted value and an actual value of a traffic flow according to still another embodiment of the present invention;
FIG. 3 is a comparison graph of a comparison mode predicted value and a traffic flow actual value in traffic flow prediction according to the present invention;
FIG. 4 is a comparison graph of a predicted value and an actual value of a traffic flow in another comparison manner when predicting the traffic flow in the present invention;
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any inventive step, are intended to be within the scope of the present invention.
Example 1
An intelligent traffic management method based on a Beidou navigation system comprises the following steps
S0, the control center is provided with a traffic flow prediction module and a Beidou navigation module, the traffic flow prediction module predicts traffic flow, the Beidou navigation module obtains the road congestion condition at the current moment, and the two modules establish one-to-one mapping;
the traffic jam condition is divided into four conditions of smooth running, slow running, jam and serious jam according to the running speed of the road, correspondingly, the detected traffic flow is also divided into four conditions to be in one-to-one correspondence, and when the two conditions are inconsistent, the abnormal traffic condition can be well found.
S1, when the two road detection units do not meet the mapping relationship, determining a plurality of road detection units which may be involved, and sending instructions to the road detection units by a control center, wherein the road detection units are provided with a Beidou satellite navigation positioning module, a communication module and a camera module which are all connected with a control module;
s2, after receiving the instructions, the control modules control the road detection unit to take a picture to acquire road picture information;
s3, the control module transmits the road photo information with position information acquired by the Beidou satellite navigation and positioning module back to the control center;
in the invention, when the traffic flow is inconsistent with the road congestion condition, the function of the road detection unit is played, so that more direct and effective real road data is obtained, and intelligent and automatic image analysis is carried out according to the traffic condition, so that the road abnormity is found by utilizing big data, the automatic image analysis in all time periods is well realized, and the calculated amount is greatly reduced.
And S4, the control center formulates a road management scheme for the road photo information. Different road jam conditions are suitable for different road management, but the jam in a reasonable range has little artificial intervention significance, so that the situation needing artificial intervention can be well targeted, the automation degree is further improved, and the artificial intervention is carried out only when the situation is suitable
Further, the road detection unit is a mobile road detection unit or a fixed road detection unit, and the mobile road detection unit is an unmanned aerial vehicle or a road driving device.
Experimental study of traffic flow prediction
A monitoring method based on dynamic traffic flow predicts traffic flow Q of one side of a crossroad to the downstream at t + m moment according to traffic conditions of the crossroad at t moment and t-m moment Prediction (t + m) Thereby carry out different level control to traffic flow, including the following step:
s1, determining the independent variable parameters of a training model as
t time three upstream traffic flows q in the direction of the crossroad 1 、q 2 、q 3 The information, obtained by video surveillance,
the crossroad at the time t and the time t-mDownstream traffic flow Q (t) 、Q (t-m) The information, obtained by video surveillance,
the downstream vehicle traffic influence level L in this direction at the intersection at time t,
one side of the crossroad downstream traffic flow Q at the moment when the dependent variable parameter is t + m Prediction (t + m) ；
Traffic flow is a dynamically changing quantity that represents the number of vehicles passing through a section of a road per unit time and that changes with time and space, in veh/h.
S2, recording historical data of the independent variable parameters and the dependent variable parameters, and performing noise reduction processing on the data through wavelet analysis;
s3, predicting to obtain a BP predicted value Q of traffic flow BP of one side of the crossroad downstream at the moment t + m through a BP neural network bp prediction (t + m) Specifically, historical data of respective variable parameters and dependent variable parameters are input into a BP neural network to obtain a BP neural network traffic model;
at the time of t, the numerical values of the respective variables are input into a BP neural network traffic model to obtain a downstream traffic flow BP predicted value Q in the direction of the crossroad at the time of t + m bp prediction (t + m) ，
S4, according to Q bp prediction (t + m) Obtaining Q Prediction (t + m) And then determining the monitoring level M of the direction of the crossroad, wherein the monitoring level M can be divided into two levels or three levels, aiming at the uncongested time period, the monitoring level with low requirement can be carried out, aiming at the congested time period, the monitoring level with high requirement can be carried out, and therefore, the monitoring resource is reasonably utilized.
Further, in step S1, the traffic influence level L of the downstream vehicle in the direction of the intersection at time t is specifically according to the following formula:
and li is from the moment t-m to the moment t, the vehicles of each passing vehicle are classified, and the cloud vehicle grades corresponding to the license plate numbers are collected by the camera to obtain.
Further, the vehicle classification li follows the following formula:
li＝a*r+b*s+c*lio，(i＝1,2,3…，n)，
wherein r is the accumulated number of roadside illegal parking in the driving process of the vehicle, s is the accumulated number of crossing at a red light in the driving process of the vehicle, lio is the vehicle classification of other vehicles under the owner name of the vehicle, a, b and c are respective weights,
in order to eliminate dimension influence among indexes, the independent variables carry out standardization processing, namely normalization processing on data of the independent variables;
in order to better utilize data, the inventor finds that whether traffic is congested or not, whether traffic flow is influenced or not, and certain factors are related to driving habits of drivers, so that the bad driving habits influence the traffic flow in a plurality of driving habits of the drivers, and the illegal parking and red light running are selected as vehicle grades through multi-party comparison, so that the vehicle grades are most representative.
And r and s are acquired by camera video acquisition.
Further, in S4, Q Prediction (t + m) ＝Q bp prediction (t + m) 。
Experiment 2
All steps included in experiment 1;
further, between S3 and S4, also include
S30, obtaining a downstream traffic flow RF predicted value Q of one side of the intersection at the t + m moment through random forest model prediction rf prediction (t + m) Specifically, historical data of respective variable parameters and dependent variable parameters are input into the random forest model to obtain a random forest traffic model;
at the time of t, the numerical values of the variables are input into the random deep forest model to obtain the downstream traffic flow RF predicted value Q in the direction of the crossroad at the time of t + m rf prediction (t + m) 。
Further, wherein, in S4, Q Prediction (t + m) ＝K 1 *Q bp prediction (t + m) +K 2 *Q rf prediction (t + m) 。
Further, in the present invention,
in S4, K 1 、K 2 Is given as initial condition K 1 +K 2 =1; and K is 1 、K 2 And calculating by an optimal weighted combination method.
Comparative example 1
A monitoring method based on dynamic traffic flow predicts traffic flow Q of one side of a crossroad to the downstream at t + m moment according to traffic conditions of the crossroad at t moment and t-m moment Prediction (t + m) So as to monitor the traffic flow at different levels, comprising the following steps:
s1, determining the independent variable parameters of a training model as
t time three upstream traffic flows q in the direction of the crossroad 1 、q 2 、q 3 ，
the downstream traffic flow Q in the direction of the crossroad at the time t and the time t-m (t) 、Q (t-m) ，
the downstream vehicle traffic influence level L in this direction at time t at the intersection,
one downstream traffic flow Q of one part of the crossroad at the moment when the dependent variable parameter is t + m Prediction (t + m) ；
S2, recording historical data of the independent variable parameters and the dependent variable parameters, and performing noise reduction processing on the data through wavelet analysis;
s30, predicting through a random forest model to obtain a downstream traffic flow RF predicted value Q of one side of the intersection at the moment t + m rf prediction (t + m) Specifically, historical data of respective variable parameters and dependent variable parameters are input into the random forest model to obtain a random forest traffic model;
at the time of t, the numerical values of the respective variables are input into the random deep forest model to obtain a downstream traffic flow RF predicted value Q in the direction of the crossroad at the time of t + m rf prediction (t + m) 。
S4 according to Q rf prediction (t + m) Obtaining Q Prediction (t + m) Then, the monitoring level M of the direction of the intersection is determined.
Further, in step S1, the downstream vehicle traffic influence level L in the direction of the intersection at time t is specifically according to the following formula:
and li is from the moment t-m to the moment t, the passing vehicles of each vehicle are classified, and the cloud vehicle grades corresponding to the license plate numbers are acquired by the camera.
Further, the vehicle classification li is according to the following formula:
li＝a*r+b*s+c*lio，(i＝1,2,3…，n)，
wherein r is the accumulated number of roadside illegal parking in the driving process of the vehicle, s is the accumulated number of crossing at a red light in the driving process of the vehicle, lio is the vehicle classification of other vehicles under the owner name of the vehicle, a, b and c are respective weights,
and r and s are acquired by camera video acquisition.
Comparative example 2
A monitoring method based on dynamic traffic flow predicts traffic flow Q of one side of a crossroad to the downstream at t + m moment according to traffic conditions of the crossroad at t moment and t-m moment Prediction (t + m) So as to monitor the traffic flow at different levels, comprising the following steps:
s1, determining the independent variable parameters of a training model as
t time three upstream traffic flows q in the direction of the crossroad 1 、q 2 、q 3 ，
the downstream traffic flow Q in the direction of the crossroad at the time t and the time t-m (t) 、Q (t-m) ，
One side of the crossroad downstream traffic flow Q at the moment when the dependent variable parameter is t + m Prediction (t + m) ；
S2, recording historical data of the independent variable parameters and the dependent variable parameters, and performing noise reduction processing on the data through wavelet analysis;
s3, predicting and obtaining one of the crossroads at the t + m moment to the downstream through a BP neural networkTraffic flow BP predicted value Q bp prediction (t + m) Specifically, historical data of respective variable parameters and dependent variable parameters are input into a BP neural network to obtain a BP neural network traffic model;
at the time of t, the numerical values of the respective variables are input into a BP neural network traffic model to obtain a downstream traffic flow BP predicted value Q in the direction of the crossroad at the time of t + m bp prediction (t + m) ，
S4, according to Q bp prediction (t + m) Obtaining Q Prediction (t + m) Then, the monitoring level M of the direction of the intersection is determined.
According to the experimental methods of experiment 1, experiment 2, comparison 1 and comparison 2, predicting and actually measuring a certain crossroad in Shijiazhuang City in Hebei province, wherein the sampling time is 6 hours to 20 hours, the sampling interval is 5 minutes, as shown in figures 1-4, fitting comparison graphs of predicted values and actual values of experiment 1, experiment 2, comparison 1 and comparison 2 are shown, the abscissa is the prediction frequency, measurement and measurement are carried out once every 5 minutes, the ordinate is the predicted value of traffic flow, the unit is veh/h, the concrete results of experiment 1, experiment 2, comparison 1 and comparison 2 are shown below,
| group of | Prediction model | Mean absolute percentage error | Root mean |
| Experiment | |||
| 1 prediction | BP neural network model 2 | 5.9％ | 11.4％ |
| Experiment 2 prediction | BP neural network model 2+ random forest model | 5.4％ | 10.6 |
| Comparison | |||
| 1 prediction | Random forest model | 7.6％ | 13.5％ |
| Contrast 2 prediction | BP |
6.2％ | 12.5％ |
As can be seen from the above table, in the experiment 1 of the present invention, compared with the comparison 2, after the influence level L of the downstream vehicle traffic in the direction of the intersection at the time t is increased by the BP neural network model 2 of the present invention, the average absolute percentage error and the root mean square error are both improved to a certain extent, and as can be seen from the comparison between the experiment 2 and the comparison 1 and the experiment 1, after the two prediction models are combined, the average absolute percentage error and the root mean square error are both further improved to a certain extent, so that the BP neural network and the random forest model algorithm are combined by the optimal weighting rule, so that the prediction effect of the combined model is improved, and is superior to the random forest algorithm and the BP neural network algorithm of the single model.
The present invention is not limited to the above preferred embodiments, and any modifications, equivalent substitutions, improvements, etc. within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (1)
1. An intelligent traffic management method based on a Beidou navigation system is characterized by comprising the following steps
S0, the control center is provided with a traffic flow prediction module and a Beidou navigation module, the traffic flow prediction module predicts traffic flow, the Beidou navigation module obtains road congestion conditions at the current moment, and one-to-one mapping is established between the traffic flow prediction module and the Beidou navigation module;
s1, when the two detection units do not meet the mapping relationship, determining a plurality of road detection units which may be involved, and sending instructions to the road detection units by a control center, wherein the road detection units are provided with a Beidou satellite navigation positioning module, a communication module and a camera module which are all connected with a control module;
s2, after receiving the instructions, the control modules control the road detection unit to take a picture to acquire road picture information;
s3, the control module transmits the road photo information with position information acquired by the Beidou satellite navigation and positioning module back to the control center;
s4, the control center formulates a road management scheme for the road photo information;
in S0, the traffic flow prediction method predicts the traffic flow Q downstream of one of the crossroads at the t + m moment according to the traffic conditions of one crossroad at the t moment and the t-m moment Prediction (t + m) Therefore, different levels of monitoring of traffic flow are performed, and the method specifically comprises the following steps:
s01, determining the independent variable parameters of the training model as
t time three upstream traffic flows q in the direction of the crossroad 1 、q 2 、q 3 The information, obtained by video surveillance,
the downstream traffic flow Q in the direction of the crossroad at the time t and the time t-m （t） 、Q （t-m） The information, obtained by video surveillance,
the downstream vehicle traffic influence level L in this direction at time t at the intersection,
one side of the crossroad downstream traffic flow Q at the moment when the dependent variable parameter is t + m Prediction (t + m) ；
S02, recording historical data of the independent variable parameters and the dependent variable parameters;
s03, predicting to obtain a BP predicted value Q of traffic flow BP of one side of the crossroad downstream at the moment t + m through a BP neural network bp prediction (t + m) Specifically, historical data of respective variable parameters and dependent variable parameters are input into a BP neural network to obtain a BP neural network traffic model;
at the time of t, the numerical values of the respective variables are input into a BP neural network traffic model to obtain a downstream traffic flow BP predicted value Q in the direction of the crossroad at the time of t + m bp prediction (t + m) ；
In step S01, the downstream vehicle traffic influence level L in the direction of the intersection at time t is specifically according to the following formula:
L= 
li/m，（i=1,2,3…，n）
the vehicle grade of each passing vehicle is acquired by acquiring the cloud vehicle grade corresponding to the license plate number through the camera from the moment t-m to the moment t;
wherein the vehicle classification li is according to the following formula:
li =a*r+b*s+c*lio，（i=1,2,3…，n），
wherein r is the accumulated number of roadside illegal parking in the driving process of the vehicle, s is the accumulated number of crossing at a red light in the driving process of the vehicle, lio is the vehicle classification of other vehicles under the owner name of the vehicle, a, b and c are respective weights,
wherein, r and s are acquired by camera video acquisition;
wherein, between S03 and S04 also includes
S030, predicting to obtain an RF predicted value Q of traffic flow of one side of the crossroad to the downstream at the t + m moment through a random forest model rf prediction (t + m) Specifically, historical data of respective variable parameters and dependent variable parameters are input into the random forest model to obtain a random forest traffic model;
at the time of t, the numerical values of the variables are input into the random deep forest model to obtain the downstream traffic flow RF forecast of the crossroad in the direction of t + m at the time of tMeasured value Q rf prediction (t + m) ；
Wherein, in S04, Q Prediction (t + m) =K 1 *Q bp prediction (t + m) + K 2 * Q rf prediction (t + m) ，
Wherein Q is Prediction (t + m) Determining the monitoring level M of the direction of the crossroad, dividing the monitoring level M into two levels or three levels, aiming at the monitoring level with low requirement in the uncongested time period and aiming at the monitoring level with high requirement in the congested time period; the road detection unit is a mobile road detection unit or a fixed road detection unit, and the mobile road detection unit is an unmanned aerial vehicle or a road running device; in S04, K 1 、K 2 Is given as initial condition K 1 +K 2 =1; and K is 1 、K 2 And calculating by an optimal weighted combination method.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011623294.5A CN112802335B (en) | 2020-12-31 | 2020-12-31 | Intelligent traffic management method based on Beidou navigation system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011623294.5A CN112802335B (en) | 2020-12-31 | 2020-12-31 | Intelligent traffic management method based on Beidou navigation system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112802335A CN112802335A (en) | 2021-05-14 |
| CN112802335B true CN112802335B (en) | 2023-04-07 |
Family
ID=75807497
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011623294.5A Active CN112802335B (en) | 2020-12-31 | 2020-12-31 | Intelligent traffic management method based on Beidou navigation system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112802335B (en) |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2018103313A1 (en) * | 2016-12-06 | 2018-06-14 | 杭州海康威视数字技术股份有限公司 | Traffic accident occurrence risk prediction method, device and system |
| CN110827543A (en) * | 2019-11-11 | 2020-02-21 | 重庆邮电大学 | Short-term traffic flow control method based on deep learning and spatio-temporal data fusion |
Family Cites Families (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7698055B2 (en) * | 2004-11-16 | 2010-04-13 | Microsoft Corporation | Traffic forecasting employing modeling and analysis of probabilistic interdependencies and contextual data |
| JP4950596B2 (en) * | 2006-08-18 | 2012-06-13 | クラリオン株式会社 | Predicted traffic information generation method, predicted traffic information generation device, and traffic information display terminal |
| CN102903237B (en) * | 2011-07-25 | 2015-06-24 | 日立(中国)研究开发有限公司 | Device and method for traffic management service |
| US9569960B2 (en) * | 2015-02-24 | 2017-02-14 | Here Global B.V. | Method and apparatus for providing traffic jam detection and prediction |
| CN106781464A (en) * | 2016-12-06 | 2017-05-31 | 广州市科恩电脑有限公司 | A kind of congestion in road situation method of testing |
| US10515542B2 (en) * | 2017-03-27 | 2019-12-24 | Iheartmedia Management Services, Inc. | Automated traffic data validation |
| CN107832890A (en) * | 2017-11-15 | 2018-03-23 | 浙江大华技术股份有限公司 | A kind of congestion Forecasting Methodology and device |
| CN111462484B (en) * | 2020-03-31 | 2021-11-30 | 腾讯科技（深圳）有限公司 | Congestion state determination method, device, equipment and computer readable storage medium |
-
2020
- 2020-12-31 CN CN202011623294.5A patent/CN112802335B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2018103313A1 (en) * | 2016-12-06 | 2018-06-14 | 杭州海康威视数字技术股份有限公司 | Traffic accident occurrence risk prediction method, device and system |
| CN110827543A (en) * | 2019-11-11 | 2020-02-21 | 重庆邮电大学 | Short-term traffic flow control method based on deep learning and spatio-temporal data fusion |
Non-Patent Citations (2)
| Title |
|---|
| 城市道路交通单向拥堵点预测分流仿真;孙冰;《计算机仿真》;20171215(第12期);全文 * |
| 基于BP神经网络的城市交通拥堵情况预测;胡鑫泽等;《电子技术与软件工程》;20200115(第01期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112802335A (en) | 2021-05-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109416873B (en) | Autonomous or partially autonomous motor vehicle with automated risk control system and corresponding method | |
| CN112071061B (en) | Vehicle service system based on cloud computing and data analysis | |
| WO2018064931A1 (en) | Method for estimating travel time distribution of taxi on urban roads when operating states of taxis are considered | |
| US11170639B2 (en) | Transportation threat detection system | |
| CN110733507B (en) | Lane changing and road isolating method, device, equipment and storage medium | |
| EP3403219A1 (en) | Driver behavior monitoring | |
| WO2016056568A1 (en) | Road condition identification system | |
| US20150045986A1 (en) | Systems and Methods for Determining Driver Fatigue Level from Lane Variability and Geographic Location | |
| CN111695418A (en) | Method and system for safe driving based on road condition detection | |
| CN111452799A (en) | Driving behavior evaluation method and system | |
| CN112712714A (en) | Traffic light timing optimization method and simulation system based on bayonet monitoring equipment | |
| CN111613070A (en) | Traffic signal lamp control method, traffic signal lamp control device, electronic equipment and computer storage medium | |
| CN113112789B (en) | Influence prediction and control method for urban expressway emergency | |
| CN112542049A (en) | Comprehensive management and control platform for intelligent traffic | |
| JP7051424B2 (en) | Anomaly detection device | |
| CN112802335B (en) | Intelligent traffic management method based on Beidou navigation system | |
| CN111767644B (en) | Method for estimating actual traffic capacity of expressway road section by considering speed limit influence of single tunnel | |
| CN110276951A (en) | One kind being based on mobile Internet Congestion prediction method | |
| CN115731693A (en) | Lane dividing method and related device | |
| CN114023065A (en) | Algorithm for intelligently diagnosing intersection service level by utilizing video analysis data | |
| CN113971885B (en) | Vehicle speed prediction method, device and system | |
| CN115909752B (en) | Method for identifying and counting sharp turns based on historical data of vehicle users | |
| CN115273468B (en) | Traffic jam control strategy generation method and device | |
| CN112562336B (en) | Real-time dynamic allocation method and system for intersection traffic signal duration | |
| CN114312796A (en) | Vehicle speed control method, vehicle speed control device, storage medium and vehicle speed control system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |