CN115100847A - Queuing service time estimation method facing low-permeability network connection vehicle trajectory data - Google Patents
Queuing service time estimation method facing low-permeability network connection vehicle trajectory data Download PDFInfo
- Publication number
- CN115100847A CN115100847A CN202210550566.6A CN202210550566A CN115100847A CN 115100847 A CN115100847 A CN 115100847A CN 202210550566 A CN202210550566 A CN 202210550566A CN 115100847 A CN115100847 A CN 115100847A
- Authority
- CN
- China
- Prior art keywords
- vehicle
- period
- service time
- queuing service
- queuing
- 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
Classifications
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
-
- G—PHYSICS
- 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
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/10—Internal combustion engine [ICE] based vehicles
- Y02T10/40—Engine management systems
Landscapes
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Traffic Control Systems (AREA)
Abstract
The invention discloses a queuing service time estimation method facing low-permeability network connection vehicle track data. Then, the oversaturated vehicle is identified and marked by judging the consistency of the arrival and departure periods of the vehicle. And then, aiming at the two conditions of undersaturation and supersaturation, converting the vehicle parking state, and constructing a vehicle queuing service time probability model based on Logistic regression by taking the converted vehicle parking state as an input. And finally, realizing dynamic estimation of queuing service time by adopting a Laplace approximation method based on Bayesian prior-posterior rolling. The method is suitable for the low-permeability data environment, can realize the queuing service time estimation under the undersaturation and supersaturation scenes under the condition of limited vehicle track samples of the internet vehicles, and can provide support for the signal control optimization based on the internet vehicle data.
Description
Technical Field
The invention relates to a queuing service time estimation method facing low-permeability network connection vehicle track data, and belongs to the technical field of queuing state estimation.
Background
The low-efficiency operation area of the urban traffic system is mainly concentrated on urban road intersections, so that the research on the traffic operation state of the intersections is particularly important. The dynamic queuing state information is used as an evaluation index of the running state of the intersection and feedback information of signal control optimization, and plays an important role in signal optimization timing and coordination control. For the research on the queuing state, the conventional method mostly utilizes fixed detector data such as coil detector data, however, the coverage rate of the fixed detector is low due to the high installation and maintenance cost of the fixed detector, and it is difficult to effectively support the traffic control application at the road network level. In order to overcome the deficiency of data space coverage, researchers at home and abroad try to research by using floating car data, mobile phone signaling data, satellite image data and unmanned aerial vehicle shooting data. In recent years, due to the development of unmanned driving technology and vehicle-road coordination, more and more vehicles such as networked cars and networked cars can provide driving track records, and the queuing state estimation problem based on the networked car track data becomes a hot spot of current research.
At present, most of researches on queuing states are carried out from the perspective of vehicle aggregation, and the queuing length or the number of queued vehicles is taken as an estimation object and is estimated based on a traffic flow theory or probability statistics. The estimation method based on the traffic flow theory takes the queuing length estimation as a deterministic process, and can be further divided into a method based on a shock wave theory and a method based on input-output according to a modeling thought. The method based on the probability statistics usually obtains the parameters of the arrival distribution according to the arrival distribution of the vehicles and the vehicle track data collected in the period, thereby obtaining the probability distribution of the queuing length, and taking the expected value as the estimated value of the queuing length. These studies have achieved certain results, but most of the methods used have an assumption about the arrival mode of the vehicle, for example, (1) in order to implement traffic wave fitting, a queuing length estimation method based on the shockwave theory, the intersection point between two shockwaves is sensitive to estimation value errors, and queuing length estimation is difficult to implement in the case of periodic data loss. In addition, the method takes the distance from the intersection point of the queue forming wave and the evanescent wave to the stop line as the queue length, and has large errors, especially under the scenes of uneven arrival modes and low saturation. (2) The input-output-based queue length estimation method requires simultaneous upstream and downstream arrangement of detectors, and the estimation accuracy is highly sensitive to the measurement error of the detectors. (3) The queuing length estimation method based on the probability statistics does not consider the uncertainty of the arrival mode of the vehicle, mostly adopts linear or piecewise linear assumption, and does not consider the arrival volatility generated by the upstream interference of the fleet. In addition, the method assumes a given constant permeability, and if permeability varies significantly over time, or is difficult to estimate accurately, the queue length estimation may also be inaccurate. Most of the existing methods have the hypothesis of vehicle arrival modes, the hypothesis may not be consistent with the actual traffic flow running condition, the estimation accuracy is easily influenced by the random fluctuation of vehicle arrival, and the accuracy and the applicability of the methods are reduced. On the other hand, the limited number of samples currently available for observation is insufficient, and most methods perform poorly at low permeability.
Although the intelligent internet technology is developing rapidly, the popularity of internet vehicles is low in the coming decade. The mixed running state of the internet vehicles and the traditional vehicles is a necessary stage, the permeability of the internet vehicles can be maintained at a lower level for a long time, and the low-permeability data environment can limit the estimation effect of the existing research method. And the partial queuing length estimation method is researched by combining with rapidly developed vehicle track data according to the model characteristics. The vehicle trajectory data required by these methods are generally high in permeability, and the methods are not good in performance in a low-permeability data environment. In order to solve the problem of insufficient vehicle track information in a low-permeability data environment, some methods adopt historical data or collect multiple periodic data, and although estimation accuracy can be improved, real-time performance cannot be guaranteed. There are also some methods that require ensuring the stability of the vehicle permeability observable in the traffic stream when queuing state studies are performed, and the methods are limited when the vehicle arrival volatility is high.
Disclosure of Invention
The technical problem to be solved by the invention is as follows: the method for estimating the queuing service time facing to the low-permeability network vehicle track data is provided, and from the vehicle dissipation angle, the intersection queuing state characterized by the queuing service time is provided, so that the assumption of a traffic arrival mode is avoided, and the estimation method under the low-permeability condition is provided.
The invention adopts the following technical scheme for solving the technical problems:
the queuing service time estimation method facing the low-permeability Internet vehicle track data comprises the following steps:
and 6, considering the relevance of traffic flow running states between adjacent periods, realizing dynamic estimation of queuing service time based on Bayesian prior-posterior rolling, taking posterior estimation information obtained from the previous period in the two adjacent periods as prior estimation information of the next period, realizing the dynamic rolling of the prior-posterior information by utilizing a Laplace approximation method in Bayesian Logistic regression, obtaining the distribution of model parameters, thus obtaining the distribution of the queuing service time, and taking the average value to finish the dynamic estimation of the periodic queuing service time.
As a preferable aspect of the present invention, in the step 2, a low-speed traveling state is set when the speed of the vehicle is less than 3m/s, it is determined whether the vehicle is in the low-speed traveling state for 5 or more consecutive seconds when the speed of the vehicle is less than 3m/s, and the vehicle is marked as a parked vehicle when the vehicle is in the low-speed traveling state for 5 or more consecutive seconds.
As a preferable aspect of the present invention, in the step 3, the vehicle arrival period is defined as a period corresponding to a time when the vehicle travels away from the stop line at a free flow speed in an ideal state when the vehicle is not influenced by the signal control and the preceding vehicle; the vehicle driving-off period is defined as a period corresponding to the actual stop line driving-off time of the vehicle;
1) determining a vehicle driving-off period according to the time of the vehicle actually driving off the stop line, selecting two track points which are nearest to the front and the back of the stop line in the vehicle track data, and determining the time of the vehicle actually driving off the stop line by using a linear interpolation method:
wherein l b 、l a Respectively showing the positions of two continuous track points in front and back of the stop line in the advancing direction of the vehicle, t b 、t a Respectively representing the time of the vehicle passing two successive track points before and after the stop line, l sb Indicating the position of the stop-line; obtaining a vehicle driving-off period by matching the actual vehicle driving-off stop line time with the signal timing;
2) acquiring the instantaneous speed of the vehicle according to the vehicle track data, establishing a ternary Gaussian mixture model, solving parameters of the ternary Gaussian mixture model by using a maximum expectation algorithm, dividing the running state of the vehicle by taking the instantaneous speed of the vehicle as the input of the ternary Gaussian mixture model based on the solved parameters, estimating the free flow speed of the vehicle, and determining the arrival period of the vehicle by combining the vehicle track data;
the vehicle running state is divided into a queuing parking and starting state, a deceleration state and a free running state, the three states correspond to three Gaussian distributions, and when the instantaneous speed x belongs to x k Posterior probability P of class>At 0.6, x is considered to belong to x k Class distribution, k is 1,2, 3.
As a preferable aspect of the present invention, in step 5, the vehicle queuing and parking state information is converted for vehicles in the undersaturation period and the supersaturation period, specifically: the parked and unmarked vehicles are represented in the model by 0 and 1, respectively, while the parked and unmarked vehicles in the next cycle of the oversaturation cycle are represented by 0 and 0, respectively;
according to the formula of Sigmoid function, the parking probability p (t) of the vehicle at the time t is:
wherein, alpha and beta are parameters of Sigmoid function;
for an undersaturated period, the queuing service time QST for that period is:
wherein G is the green time of the undersaturation period;
for the supersaturation period, vehicles which arrive at the e period and leave at the e + N period exist, N is a positive integer, and the queuing service time of the e period is as follows:
wherein G is i Is the green light time of the i-cycle,
representing all green light times from the e-period to the e + N-1 period.
Compared with the prior art, the invention adopting the technical scheme has the following technical effects:
1. the invention considers the vehicle queuing dissipation process, selects the queuing service time as the queuing state representation index, avoids the assumption of the vehicle arrival mode, overcomes the defects and the limitations brought by the arrival model assumption, better accords with the actual situation and has better universality.
2. The invention considers the relevance of the queuing state of the intersection on the continuous period, adopts a Laplace approximation method, constructs a queuing service time dynamic estimation method based on Bayesian, and realizes the periodic queuing service time dynamic estimation.
3. The periodic queuing service time dynamic estimation method based on the Bayesian Logistic method is suitable for the low-permeability real environment of the current internet vehicle, is also suitable for under-saturated and over-saturated scenes, and has good estimation precision.
Drawings
FIG. 1 is a flow chart of a queuing service time estimation method facing low-permeability internet connection trajectory data according to the invention;
FIG. 2 is a plot of log velocity after Gaussian distribution fitting;
FIG. 3 is an estimation result of a Bayesian Logistic regression method under different permeability data environments; wherein (a) is 6% permeability, (b) is 10% permeability, (c) is 20% permeability, and (d) is 30% permeability.
Detailed Description
Reference will now be made in detail to embodiments of the present invention, examples of which are illustrated in the accompanying drawings. The embodiments described below with reference to the accompanying drawings are illustrative only for the purpose of explaining the present invention, and are not to be construed as limiting the present invention.
Interpretation of terms:
the internet vehicle integrates a V2X communication technology, and intelligent information exchange and sharing with people, vehicles, roads, cloud ends and the like can be realized through the vehicle-mounted sensing system and the information terminal.
And low-permeability vehicle track data, wherein the low permeability refers to that the permeability is lower than 30%, and the vehicle track data refers to information such as vehicle numbers, track point time, vehicle positions, vehicle speeds and the like.
The queuing service time is the time for dissipating the queued vehicles within the green light time, and is defined as the green light time between the green light turning-on time of a certain period and the time when the last parked vehicle arrives at the period and leaves the stop line.
The vehicle arrival period is defined as the period during which the vehicle travels away from the stop line at the free stream speed in an ideal state when it is not affected by the signal control and the preceding vehicle.
The vehicle drive-off period is defined as a period corresponding to the actual stop-line drive-off time of the vehicle.
The Logistic model is used for modeling the probability of existence of a certain category or event, and can also be extended to modeling several types of events.
As shown in fig. 1, from the perspective of vehicle dissipation, the present invention proposes an intersection queuing state characterized by queuing service time, avoids the assumption of traffic arrival mode, and provides an estimation method under a low permeability condition, specifically comprising the following steps:
s1: and acquiring auxiliary information such as intersection internet vehicle track data and a signal timing scheme.
S2: and extracting the information of the vehicle queuing and parking state. The invention judges whether the vehicle passes through the intersection to stop in line or not according to the vehicle speed information, the invention provides that the vehicle speed is less than 3m/s and is in a low-speed driving state, and when the vehicle is in the low-speed driving state for more than 5 seconds continuously, the vehicle is marked as a stopped vehicle.
S3: and (5) carrying out consistency judgment on the arrival and departure periods of the vehicles, and identifying the oversaturated vehicles. Each vehicle is classified into a certain period according to the arrival period. For a certain period, if the arrival and departure periods of all vehicles are the same, the period is an undersaturation period; if the arrival and departure periods of the vehicle are different, the vehicle is judged to be an oversaturated vehicle, and the period is judged to be an oversaturated period.
S31: and determining a driving-off period according to the time when the vehicle drives off the stop line. Selecting the nearest front and back of the stop line in the vehicle track dataTwo track points, and the time t of the vehicle driving from the stop line is determined by utilizing a linear interpolation method p ：
Wherein l b 、l a Indicating the position of two successive track points before and after the stop line, t b 、t a Time of these two trace points, l sb Indicating the position of the stop-line. By matching the vehicle drive-off stop-line time with the signal timing, a drive-off period can be obtained.
S32: and extracting the free flow speed, and determining an arrival period by combining track data. And taking the instantaneous speed of the vehicle in the historical track data as input information, establishing a Gaussian mixture model to divide the running state of the vehicle, estimating the speed of the free flow of the vehicle in running, and determining the arrival period of the vehicle by combining the vehicle track data. The specific process of the step is as follows:
the driving states of the vehicle are mainly divided into a queuing parking and starting state, a deceleration state and a free driving state, the three states correspond to three Gaussian distributions, and the queuing parking and starting state, the deceleration state and the free driving state correspond to one another in sequence from small to large according to the mean value of the Gaussian distributions. When data x i Belong to x k Posterior probability P (x) of class i ∈k j |x i )>At 0.6, x is considered to be i Belonging to this class of distributions.
The multivariate gaussian mixture can be expressed as:
wherein X is a sample set, alpha k Is a mixed weight of the k-th Gaussian distribution and
is a probability density function of the kth gaussian distribution.
Calculating posterior probability and maximum likelihood estimation in two steps by using a maximum Expectation-Maximization algorithm (Expectation-Maximization algorithm) to obtain new parameters, and performing iterative calculation until likelihood converges.
S4: and converting the vehicle queuing and parking state information as input data of the model aiming at the two conditions of undersaturation and supersaturation. For the oversaturation period, the parking state of the vehicle in the next period needs to be changed to 0.
S5: and establishing a vehicle queuing service time probability model based on Logistic regression, wherein the queuing service time is represented by integrating the parking probability.
According to a Sigmoid function formula, the parking probability of the vehicle at the time t is assumed as follows:
wherein, α and β are parameters of Sigmoid function, and are related to the degree of steepness and translation of Sigmoid function curve.
For an undersaturation period, the queuing service time for the period is:
where G is the green time of the cycle.
Under the condition of supersaturation, for vehicles arriving at the e period and leaving at the e + N period, the estimated value of the queuing service time of the e period can be obtained only after the e + N period, in this case, all observable vehicle information in the e period to the e + N period can be used as the estimated information of the queuing service time of the e period, and the queuing service time of the e period is as follows:
wherein G is i At i period of green lightThe second half of the right side of the equation represents all green light times from the e period to the e + N-1 period.
S6: and realizing dynamic estimation of queuing service time based on Bayesian prior-posterior rolling. Considering the relevance of traffic flow running states between adjacent periods, taking the posterior estimation information obtained in the previous period as the prior estimation information of the period, realizing the dynamic rolling of the prior-posterior information by utilizing a Laplace approximation method in Bayesian Logistic regression, and obtaining parameters (alpha, beta) T And (e) obtaining the distribution of the queuing service time, and taking the average value of the distribution to finish the dynamic estimation of the periodic queuing service time.
The posterior distribution of the parameters can be obtained by a Bayesian formula:
in the formula, theta is a parameter to be solved, x is a set of information pairs of all observed vehicles in an M period, P (theta | x) is posterior distribution of the parameter, L (x | theta) is a likelihood function of theta, P (theta) is prior distribution of theta, and P (x) is a marginal density function of x.
Approximating the posterior probability distribution to a Gaussian distribution by a Laplace approximation method, and solving the Gaussian distribution approximate to the posterior probability distribution:
wherein, theta 0 Is the mean of an approximate Gaussian distribution with a posterior probability function at θ 0 Local maximum can be obtained; a is the inverse of the variance of the approximate Gaussian distribution, and for a multidimensional density function, A is lnf (theta) at the stagnation point theta 0 The blackplug matrix.
After the posterior distribution probability of the parameters is obtained in each period, the queuing service time distribution of the period can be obtained according to a queuing service time integral formula. Taking the parameter posterior distribution of the mth period as the prior distribution of the (m + 1) th period; for the first cycle, the prior probability distribution is assumed empirically. And according to the continuous iteration of the prior probability and the posterior probability, the dynamic estimation of the queuing service time can be realized.
The method adopts a Lankershirm-Boulevard data set in the NGSIM measured data, and randomly samples vehicle track data with different permeabilities from the Lankershirm-Boulevard data set to represent the Internet vehicle track data. And (3) carrying out verification evaluation on the method by matching with actual road information and timing signal schemes of the Lankershirm major road.
Regarding the extraction of the free flow speed, the present invention divides the vehicle state into three types, and therefore the number of vehicle speed distribution groups is 3. And fitting the extracted speed set through a ternary Gaussian mixture model, and performing state division on all vehicle speed information by taking the posterior probability 0.6 as a decision boundary threshold. The velocity after the natural logarithm is taken as an abscissa and the velocity concentration frequency/group spacing is taken as an ordinate, so that a ternary gaussian distribution velocity map as shown in fig. 2 can be drawn. As is apparent from fig. 2, the vehicle is sequentially divided into three states according to the vehicle speed, the speed mean values are respectively 10.6, 30.1 and 46.5, and the distribution of each state is approximately gaussian. Therefore, a free-stream travel speed of 46.5km/h can be obtained for the section.
According to the full-sample high-precision track data and the timing scheme, the real value of the queuing service time can be obtained. And inputting the vehicle track data with different permeabilities into the method to obtain the queuing service time estimated value. In order to reflect the accuracy of the estimation result, evaluation indexes MAE (mean absolute error) and MAPE (mean absolute percentage error) are introduced.
In order to compare the estimation effects of the Bayesian Logistic regression method provided by the invention under different permeabilities, vehicle trajectory data with 6%, 10%, 20% and 30% permeabilities are selected for testing. As shown in (a), (b), (c) and (d) of fig. 3, it can be found that the MAE and the MAPE are estimated to gradually decrease in the process of increasing the permeability from 6% to 30%, and the MAE is: 6.1s, 4.8s, 3.4s, 2.5s, MAPE are: 25.5%, 19.8%, 14.6%, 10.5%. This shows that with increasing permeability, the information that can be obtained is increasing, and the accuracy of the method becomes higher; meanwhile, the estimation method provided by the invention still has good precision in a low-permeability data environment.
The above embodiments are only for illustrating the technical idea of the present invention, and the protection scope of the present invention is not limited thereby, and any modifications made on the basis of the technical solution according to the technical idea of the present invention fall within the protection scope of the present invention.
Claims (4)
1. The queuing service time estimation method facing the low-permeability Internet vehicle track data is characterized by comprising the following steps of:
step 1, acquiring a signal timing scheme of an intersection and track data of a vehicle passing through the intersection;
step 2, extracting speed information of the vehicle from the vehicle track data of the internet vehicle, judging whether the vehicle is queued to stop when passing through the intersection according to the speed information of the vehicle, namely whether the vehicle has red light states such as parking and the like, if so, marking as the vehicle to stop, otherwise, not marking, thereby obtaining vehicle queuing and stopping state information;
step 3, defining a vehicle arrival period and a vehicle driving-off period, and determining the vehicle arrival period and the vehicle driving-off period;
step 4, classifying each vehicle into a corresponding period according to the arrival period of the vehicles; for each period, if the arrival period and the driving-off period of all vehicles in the period are the same, the period is an undersaturation period; if the arrival period and the driving-off period of the vehicle are different in the period, the vehicle with the arrival period and the driving-off period different is judged to be a supersaturated vehicle, and the period in which the supersaturated vehicle is located is a supersaturated period;
step 5, establishing a vehicle queuing service time probability model based on Logistic regression, converting vehicle queuing parking state information as input of the model, and expressing queuing service time as integral of parking probability;
and 6, considering the relevance of traffic flow running states between adjacent periods, realizing dynamic estimation of queuing service time based on Bayesian prior-posterior rolling, taking posterior estimation information obtained from the previous period in the two adjacent periods as prior estimation information of the next period, realizing the dynamic rolling of the prior-posterior information by utilizing a Laplace approximation method in Bayesian Logistic regression, obtaining the distribution of model parameters, thus obtaining the distribution of the queuing service time, and taking the average value to finish the dynamic estimation of the periodic queuing service time.
2. The low-permeability internet-connected vehicle trajectory data-oriented queuing service time estimation method according to claim 1, wherein in the step 2, a low-speed running state is set when the speed of the vehicle is less than 3m/s, whether the speed of the vehicle is less than 3m/s is judged, when the speed of the vehicle is less than 3m/s, whether the vehicle is in the low-speed running state for more than 5 consecutive seconds is judged, and if the vehicle is in the low-speed running state for more than 5 consecutive seconds, the vehicle is marked as a parked vehicle.
3. The low-permeability internet vehicle trajectory data-oriented queuing service time estimation method according to claim 1, wherein in the step 3, the vehicle arrival period is defined as a period corresponding to a time when the vehicle travels away from the stop line at a free flow speed under an ideal state when the vehicle is not influenced by signal control and a preceding vehicle; the vehicle driving-off period is defined as a period corresponding to the actual stop line driving-off time of the vehicle;
1) determining a vehicle driving-off period according to the time of the vehicle actually driving off the stop line, selecting two nearest track points before and after the stop line in the vehicle track data, and determining the time of the vehicle actually driving off the stop line by using a linear interpolation method:
wherein l b 、l a Respectively showing the positions of two continuous track points in front and back of the stop line in the advancing direction of the vehicle, t b 、t a Respectively representing the time of the vehicle passing two successive track points before and after the stop line, l sb Indicating the position of the stop line; by matching actual driving of vehiclesThe time of leaving the stop line is matched with the signal, so that the vehicle leaving period is obtained;
2) acquiring the instantaneous speed of the vehicle according to the vehicle track data, establishing a ternary Gaussian mixture model, solving parameters of the ternary Gaussian mixture model by using a maximum expectation algorithm, dividing the running state of the vehicle by taking the instantaneous speed of the vehicle as the input of the ternary Gaussian mixture model based on the solved parameters, estimating the free flow speed of the vehicle, and determining the arrival period of the vehicle by combining the vehicle track data;
the vehicle running state is divided into a queuing parking and starting state, a deceleration state and a free running state, the three states correspond to three Gaussian distributions, and when the instantaneous speed x belongs to x k Posterior probability P of class>At 0.6, x is considered to belong to x k Class distribution, k is 1,2, 3.
4. The low-permeability internet vehicle trajectory data-oriented queuing service time estimation method according to claim 1, wherein in the step 5, vehicle queuing parking state information is converted for vehicles in an undersaturation period and an oversaturation period, specifically: the parked and unmarked vehicles are represented in the model by 0 and 1, respectively, while the parked and unmarked vehicles in the next cycle of the oversaturation cycle are represented by 0 and 0, respectively;
according to the formula of Sigmoid function, the parking probability p (t) of the vehicle at the time t is:
wherein, alpha and beta are parameters of Sigmoid function;
for an undersaturated period, the queuing service time QST for that period is:
wherein G is the green time of the undersaturation period;
for the supersaturation period, vehicles which arrive at the e period and leave at the e + N period exist, N is a positive integer, and the queuing service time of the e period is as follows:
wherein, G i Is the green light time of the i-cycle,
representing all green lamp times from the e period to the e + N-1 period.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210550566.6A CN115100847B (en) | 2022-05-18 | 2022-05-18 | Queuing service time estimation method for low-permeability network-connected track data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210550566.6A CN115100847B (en) | 2022-05-18 | 2022-05-18 | Queuing service time estimation method for low-permeability network-connected track data |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115100847A true CN115100847A (en) | 2022-09-23 |
| CN115100847B CN115100847B (en) | 2023-05-26 |
Family
ID=83289774
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210550566.6A Active CN115100847B (en) | 2022-05-18 | 2022-05-18 | Queuing service time estimation method for low-permeability network-connected track data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115100847B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116246466A (en) * | 2023-03-13 | 2023-06-09 | 长安大学 | Hybrid traffic flow management method and system considering autopilot multimode characteristics |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106652458A (en) * | 2017-02-20 | 2017-05-10 | 东南大学 | Online urban road route travel time estimation method based on virtual vehicle locus reconstruction |
| CN108053645A (en) * | 2017-09-12 | 2018-05-18 | 同济大学 | A kind of signalized intersections cycle flow estimation method based on track data |
| CN108665704A (en) * | 2018-05-17 | 2018-10-16 | 北京航空航天大学 | Intersection period maximum queue length method of estimation based on crowdsourcing track data |
| CN110046209A (en) * | 2019-03-26 | 2019-07-23 | 太原科技大学 | A kind of track stopping point extracting method based on Gauss model |
| CN110322704A (en) * | 2019-06-14 | 2019-10-11 | 东南大学 | A kind of queue length real-time estimation method based on number plate identification and GPS data |
| CN110335459A (en) * | 2019-04-26 | 2019-10-15 | 同济大学 | The intersection queue length estimation method and device of low-permeability track of vehicle data |
| CN112530177A (en) * | 2020-11-23 | 2021-03-19 | 西南交通大学 | Kalman filtering-based vehicle queuing length estimation method in Internet of vehicles environment |
| CN113643545A (en) * | 2021-10-14 | 2021-11-12 | 华砺智行(武汉)科技有限公司 | Vehicle queuing length sensing method, device, equipment and storage medium |
| CN113947899A (en) * | 2021-09-30 | 2022-01-18 | 南京云析科技有限公司 | Dynamic estimation method for queuing service time under low-permeability track data |
| CN114141013A (en) * | 2021-11-25 | 2022-03-04 | 东南大学 | Intersection signal timing method based on low-permeability network connection track data |
-
2022
- 2022-05-18 CN CN202210550566.6A patent/CN115100847B/en active Active
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106652458A (en) * | 2017-02-20 | 2017-05-10 | 东南大学 | Online urban road route travel time estimation method based on virtual vehicle locus reconstruction |
| CN108053645A (en) * | 2017-09-12 | 2018-05-18 | 同济大学 | A kind of signalized intersections cycle flow estimation method based on track data |
| CN108665704A (en) * | 2018-05-17 | 2018-10-16 | 北京航空航天大学 | Intersection period maximum queue length method of estimation based on crowdsourcing track data |
| CN110046209A (en) * | 2019-03-26 | 2019-07-23 | 太原科技大学 | A kind of track stopping point extracting method based on Gauss model |
| CN110335459A (en) * | 2019-04-26 | 2019-10-15 | 同济大学 | The intersection queue length estimation method and device of low-permeability track of vehicle data |
| CN110322704A (en) * | 2019-06-14 | 2019-10-11 | 东南大学 | A kind of queue length real-time estimation method based on number plate identification and GPS data |
| CN112530177A (en) * | 2020-11-23 | 2021-03-19 | 西南交通大学 | Kalman filtering-based vehicle queuing length estimation method in Internet of vehicles environment |
| CN113947899A (en) * | 2021-09-30 | 2022-01-18 | 南京云析科技有限公司 | Dynamic estimation method for queuing service time under low-permeability track data |
| CN113643545A (en) * | 2021-10-14 | 2021-11-12 | 华砺智行(武汉)科技有限公司 | Vehicle queuing length sensing method, device, equipment and storage medium |
| CN114141013A (en) * | 2021-11-25 | 2022-03-04 | 东南大学 | Intersection signal timing method based on low-permeability network connection track data |
Non-Patent Citations (1)
| Title |
|---|
| 陈之强;王辛岩;: "城市交通网络客流疏散安全评估方法仿真", 计算机仿真 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116246466A (en) * | 2023-03-13 | 2023-06-09 | 长安大学 | Hybrid traffic flow management method and system considering autopilot multimode characteristics |
| CN116246466B (en) * | 2023-03-13 | 2024-01-23 | 长安大学 | Hybrid traffic flow management method and system considering autopilot multimode characteristics |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115100847B (en) | 2023-05-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106355907B (en) | Signalized crossing queue length real-time estimation method based on track of vehicle | |
| CN110610260B (en) | Driving energy consumption prediction system, method, storage medium and equipment | |
| CN110176139B (en) | DBSCAN + -based road congestion identification visualization method | |
| CN104778834B (en) | Urban road traffic jam judging method based on vehicle GPS data | |
| CN102737504B (en) | Method for estimating bus arrival time in real time based on drive characteristics | |
| CN106846804B (en) | The real-time saturation volume rate method of estimation in intersection based on hidden Markov chain | |
| CN108198425A (en) | A kind of construction method of Electric Vehicles Driving Cycle | |
| CN111898501B (en) | Unmanned aerial vehicle online aerial photography vehicle identification and statistics method for congested road sections | |
| CN110969142B (en) | Abnormal driving scene extraction method based on network-connected vehicle natural driving data | |
| CN111581538B (en) | Expressway charging data-based expressway traffic flow state deducing method | |
| CN106971546B (en) | Road section bus permeability estimation method based on bus GPS data | |
| CN112017436B (en) | Method and system for predicting urban traffic travel time | |
| CN113947899B (en) | Queuing service time dynamic estimation method under low-permeability track data | |
| CN107564290A (en) | A kind of urban road intersection saturation volume rate computational methods | |
| CN110836675A (en) | Decision tree-based automatic driving search decision method | |
| CN110070734A (en) | Signalized intersections saturation headway estimation method based on gauss hybrid models | |
| CN104750963A (en) | Intersection delay time estimation method and device | |
| CN109493449B (en) | Freight train cargo state estimation method based on freight train GPS track data and high-speed transaction data | |
| CN106297296B (en) | A kind of fine granularity hourage distribution method based on sparse track point data | |
| CN115935672A (en) | Fuel cell automobile energy consumption calculation method fusing working condition prediction information | |
| CN108399742B (en) | Traffic situation heat map visualization method based on traffic saturation | |
| CN110766940A (en) | Method for evaluating running condition of road signalized intersection | |
| CN104916124A (en) | Public bicycle system regulation and control method based on Markov model | |
| CN115100847A (en) | Queuing service time estimation method facing low-permeability network connection vehicle trajectory data | |
| CN103903432A (en) | Equipment for determining road link congestion state and method |
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 |