CN115311844A - Highway traffic state estimation method based on self-supervision learning support vector machine - Google Patents

Highway traffic state estimation method based on self-supervision learning support vector machine Download PDF

Info

Publication number
CN115311844A
CN115311844A CN202210714620.6A CN202210714620A CN115311844A CN 115311844 A CN115311844 A CN 115311844A CN 202210714620 A CN202210714620 A CN 202210714620A CN 115311844 A CN115311844 A CN 115311844A
Authority
CN
China
Prior art keywords
traffic
traffic state
highway
traffic data
classifier
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
Application number
CN202210714620.6A
Other languages
Chinese (zh)
Other versions
CN115311844B (en
Inventor
赵妍
芮一康
冉斌
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Southeast University
Original Assignee
Southeast University
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Southeast University filed Critical Southeast University
Priority to CN202210714620.6A priority Critical patent/CN115311844B/en
Publication of CN115311844A publication Critical patent/CN115311844A/en
Application granted granted Critical
Publication of CN115311844B publication Critical patent/CN115311844B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

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
    • G08G1/0133—Traffic data processing for classifying traffic situation
    • 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 highway traffic state estimation method based on an automatic supervision learning support vector machine, which comprises the following steps: generating a training traffic data set; constructing a traffic state classification model based on an automatic supervision learning support vector machine, and training the traffic state classification model by using a traffic data characteristic vector of a highway sample; solving the decision weight parameter of the classifier to obtain different traffic states and the decision weight parameter of the corresponding vector classifier; and estimating the traffic state of the expressway at the measured time period by calculating the interval distance between the traffic data of the expressway at the measured time period and the traffic state classifier. The invention realizes the estimation of the traffic state of the highway, thereby facilitating the highway manager to carry out the pertinence control and evaluation on the traffic operation of the highway and improving the overall performance of the traffic of the road network.

Description

Highway traffic state estimation method based on self-supervision learning support vector machine
Technical Field
The invention belongs to the technical field of intelligent traffic systems and traffic state estimation, and particularly relates to a highway traffic state estimation method based on an automatic supervision learning support vector machine.
Background
With the increasing expansion of the scale of the highway network in China, the real-time supervision and management of the traffic operation condition of the highway also needs more information technology support. At present, aiming at accidental traffic congestion or accidents caused by emergencies, the main solution is to acquire information in a manual alarm manner and further take management and rescue measures; and the common traffic congestion caused by the contradiction between supply and demand is allowed to dissipate by itself. The reason for this is that traffic travelers, managers, and planners do not have comprehensive knowledge of traffic status information. Therefore, the management of the traffic state of the highway should require further enhancement of the level of services in terms of informatization, intelligence, and humanization.
The traffic state estimation of the expressway is an important component of an intelligent traffic system and is the basis of an intelligent traffic control, guidance and coordination system. Generally, the estimation of the traffic state of the highway is realized by analyzing various traffic data through various discrimination algorithms on the basis of real-time acquisition of the traffic data of the highway, comparing the analyzed traffic data with a priori traffic state standard to obtain the operation state of the current traffic system, and realizing intelligent control, management and induction of the traffic system according to the discrimination result. The method realizes quick and effective traffic state estimation, and is an important guarantee for realizing real-time intelligent and effective control of highway traffic.
In intelligent traffic management systems, various discriminant algorithms are often used to make traffic state estimates. However, at present, there is no method for solving the classification problem of the high-dimensional traffic data features based on massive unlabeled data.
Disclosure of Invention
The technical problem to be solved is as follows: the invention discloses a highway traffic state estimation method based on an automatic supervision learning support vector machine, which mainly calculates and deduces three state variables of traffic flow, interval average speed and space occupancy, considers traffic multidimensional characteristics to estimate the traffic state, and realizes the highway traffic state estimation based on the automatic supervision learning support vector machine, thereby facilitating a highway manager to carry out targeted control and evaluation on highway traffic operation and improving the overall performance of road network traffic.
The technical scheme is as follows:
a highway traffic state estimation method based on an automatic supervision learning support vector machine comprises the following steps:
s1: extracting a traffic data characteristic vector of a highway sample, wherein the traffic data characteristic vector of the highway sample comprises flow, interval average speed and occupancy; constructing a traffic state label prejudge device based on self-supervision learning, and performing self-supervision pre-training on traffic state labels on input highway sample traffic data to generate traffic state soft labels; generating a training traffic data set;
s2: constructing a traffic state classification model based on an automatic supervision learning support vector machine, and training the traffic state classification model by using a traffic data characteristic vector of a highway sample;
s3: solving the decision weight parameter of the classifier to obtain different traffic states and the decision weight parameter of the corresponding vector classifier;
s4: and estimating the traffic state of the expressway at the measured time period by calculating the interval distance between the traffic data of the expressway at the measured time period and the traffic state classifier.
Further, in step S1, the process of generating the training traffic data set includes the following steps:
s11, selecting the flow, the interval average speed and the occupancy as traffic data feature vectors of the highway samples, and performing normalization processing, wherein the normalization processing specifically comprises the following steps:
Figure BDA0003707362800000021
in the formula (I), the compound is shown in the specification,
Figure BDA0003707362800000022
the traffic data feature vector is a highway sample traffic data feature vector, comprising three dimensions,
Figure BDA0003707362800000023
the traffic data is average speed traffic data in an interval of the highway, and o is the space occupancy of the highway;
s12, constructing a traffic state label prejudge device based on a simple self-training thought in self-supervision learning, and performing self-supervision pre-training on traffic state labels on input highway sample traffic data to generate traffic state soft labels; wherein, the ratio of the average speed to the free flow rate in the selected interval is used as the identification index of the traffic state label prejudger:
Figure BDA0003707362800000024
in the formula (I), the compound is shown in the specification,
Figure BDA0003707362800000025
for the soft tag of the traffic status for the period j,
Figure BDA0003707362800000026
is average speed traffic data of j time interval v free Is freeway flow rate;
s13, generating a training traffic data set, specifically:
Figure BDA0003707362800000027
wherein T is a training traffic data set,
Figure BDA0003707362800000028
and j =1, 2.. And s, s is the total number of samples for the characteristic vector of the expressway sample traffic data in the j period.
Further, in step S2, the traffic state classification model
Figure BDA0003707362800000029
Comprises the following steps:
Figure BDA00037073628000000210
Figure BDA00037073628000000211
in the formula, sgn represents a sign function, w k Decision of hyperplane weights for class k, b k For the bias value of the decision function,
Figure BDA00037073628000000212
is a j-period expressway sample traffic data feature vector,
Figure BDA0003707362800000031
for the j-slot traffic status soft tag, j =1, 2.
Further, in step S3, solving the classifier decision weight parameters to obtain different traffic states and corresponding vector classifier decision weight parameters includes the following steps:
s31, making a decision weight parameter w of the traffic state classifier k ，b k The solution problem is converted into solution | | | w k And | | l minimization, according to the support vector space constraint, setting the constraint expression of the support vector space constraint as affine function constraint and further refining the affine function constraint as a convex optimization problem under the affine function constraint, wherein the specific formula is as follows:
Figure BDA0003707362800000032
Figure BDA0003707362800000033
in the formula, epsilon k The allowable range of error for a particular sample point for the class k decision hyperplane, C is the error cost coefficient,
Figure BDA0003707362800000034
as a relaxation variable, g (w) k ，b k ) Denotes w k And b k A constraint condition between;
s32, solving the constraint expression by using a Lagrange multiplier method, wherein the specific formula is as follows:
Figure BDA0003707362800000035
in the formula, λ j Is the corresponding lagrange multiplier;
s33, calculating the partial derivatives of the variables to make the values of the variables be 0, wherein the specific formula is as follows:
Figure BDA0003707362800000036
Figure BDA0003707362800000037
Figure BDA0003707362800000038
Figure BDA0003707362800000039
and (3) pushing out:
Figure BDA00037073628000000310
further, in step S3, the process of solving the classifier decision weight parameters to obtain different traffic states and corresponding vector classifier decision weight parameters further includes the following steps:
s34: converting the solving problem of the constraint expression into a corresponding dual problem, wherein the specific formula is as follows:
Figure BDA0003707362800000041
s.t.λ j ≥0
s35: the Gaussian kernel function is used for carrying out the upscaling conversion model, and the specific formula is as follows:
Figure BDA0003707362800000042
s.t.λ j ≥0
in the formula (I), the compound is shown in the specification,
Figure BDA0003707362800000043
is a kernel function; delta is the width of the Gaussian kernel function neuron; q (lambda) j ) Is λ j The solution function of (2); lambda [ alpha ] m Is a function of the lagrange multiplier and is,
Figure BDA0003707362800000044
is a soft label of the traffic state in the period of m,
Figure BDA0003707362800000045
for a highway sample traffic data feature vector at m time intervals, m =1,2,. The, s, j is not equal to m;
s36: by solving for λ i To obtain an optimal solution w k * And b k * The concrete formula is as follows:
Figure BDA0003707362800000046
further, in step S4, the traffic state of the expressway at the measured period is estimated by calculating the distance between the traffic data at the measured period and the traffic state classifier on the expressway with the following formula:
Figure BDA0003707362800000047
has the beneficial effects that:
firstly, the method for estimating the traffic state of the expressway based on the Self-Supervised Learning support vector machine carries out Self-Supervised pre-training based on a simple Self-training idea (simple Self-training) in Self-Supervised Learning (Self-Supervised Learning), automatically constructs Supervised soft label data from unsupervised data, and learns a pre-training model. Based on massive non-label data, a strong general representation model is obtained by learning through self-supervision learning. In the upper-layer operation, the target of a specific task is realized based on supervised learning, and intelligent control is realized based on reinforcement learning. The self-supervised learning can enhance the performance and generalization capability of the supervised learning and the reinforcement learning model. The unsupervised data is better utilized, and the internal information of the input unsupervised data can be learned. The effect of follow-up supervision learning task is promoted.
Secondly, the method for estimating the traffic state of the expressway based on the self-supervision learning support vector machine designs the estimation of the traffic state of the expressway based on the support vector machine, and effectively solves the problem of classification of high-dimensional traffic data characteristics. The obtained decision hyperplane memory occupies less space, and a kernel function is applied, so that the problem of nonlinear classification is flexibly solved. Meanwhile, the method has clear calculation flow and simple calculation, and has higher calculation speed on the premise of ensuring the accuracy of traffic state estimation.
Drawings
FIG. 1 is a flow chart of a method for estimating a traffic state of a highway based on an automatic supervised learning support vector machine according to an embodiment of the present invention;
fig. 2 is a schematic flow chart of generating a training traffic data set according to an embodiment of the present invention.
Fig. 3 is a schematic flow chart of constructing a traffic state classification model based on an unsupervised learning support vector machine according to an embodiment of the present invention.
Fig. 4 is a schematic flow chart illustrating a process of solving a classifier decision weight parameter according to an embodiment of the present invention.
Fig. 5 is a schematic diagram of a traffic state estimation result according to an embodiment of the present invention.
Detailed Description
The following examples will give the skilled person a more complete understanding of the present invention, but do not limit the invention in any way.
Referring to fig. 1, the present embodiment discloses a method for estimating a traffic state of a highway based on an auto-supervised learning support vector machine, where the method for estimating a traffic state of a highway includes the following steps:
s1: extracting a traffic data characteristic vector of a highway sample, wherein the traffic data characteristic vector of the highway sample comprises flow, interval average speed and occupancy; constructing a traffic state label prejudge device based on self-supervision learning, and performing self-supervision pre-training on traffic state labels on input highway sample traffic data to generate traffic state soft labels; a training traffic data set is generated.
S2: and constructing a traffic state classification model based on an automatic supervision learning support vector machine, and training the traffic state classification model by using the traffic data characteristic vector of the highway sample.
S3: and solving the decision weight parameters of the classifier to obtain different traffic states and the decision weight parameters of the corresponding vector classifier.
S4: and estimating the traffic state of the expressway at the measured period by calculating the distance between the traffic data of the expressway at the measured period and the traffic state classifier.
S1: extracting traffic data characteristic vectors of highway samples to generate a training traffic data set;
in an alternative embodiment of the present invention, as shown in fig. 2, step S1 specifically includes the following substeps S1-1 to S1-3:
in an optional embodiment of the invention, the method comprises the step S1-1 of selecting the flow, the interval average speed and the occupancy as the characteristic vectors of the highway sample traffic data and carrying out normalization processing. Specifically, a basic road section (about 7 kilometers in length) of a certain highway is selected, a plurality of traffic detectors are arranged in the road section, and required traffic parameters including traffic volume, interval average speed and space occupancy are provided. In the highway traffic state estimation, 7-day (24 hours in whole day, 5 minutes in detection time step) detection data is selected as a training set, and 3-day detection data is selected as a test set.
In an optional embodiment of the present invention, the present invention selects a ratio of an average vehicle speed to a free flow rate in an interval as a traffic status label pre-judging device identification index in steps S1-2, and the specific formula is as follows:
Figure BDA0003707362800000061
in the formula (I), the compound is shown in the specification,
Figure BDA0003707362800000062
for the soft tag of the traffic status for the period j,
Figure BDA0003707362800000063
is average speed traffic data of j time interval v free Is highway free flow. The traffic state of the road section is divided into four categories according to the selected high-speed actual traffic condition, and the grade division is shown in the following table 1. It should be understood that the identifiers can be defined by themselves, and the four identifiers of the present embodiment are only for illustration; the division criterion of the fluidity index can also be adjusted according to the actual scene, and is not limited to the one in table 1.
TABLE 1
Figure BDA0003707362800000064
In an optional embodiment of the present invention, the present invention generates a training traffic data set in steps S1-3, specifically:
Figure BDA0003707362800000065
wherein T is a training traffic data set,
Figure BDA0003707362800000066
and (4) obtaining a j-period expressway sample traffic data feature vector, wherein s is the total sample number.
S2: training a traffic state classification model by using the traffic data feature vector of the highway sample;
in an optional embodiment of the present invention, as shown in fig. 3, step S2 is to train a traffic state classification model by using a traffic data feature vector of a highway sample, and solve a classifier decision weight parameter to obtain different traffic states and a corresponding vector classifier decision weight parameter, and specifically includes the following substeps S2-1 to S2-4:
in an optional embodiment of the present invention, the present invention utilizes the step S1 to extract the traffic data feature vector of the highway sample to train the traffic state classification model in the step S2-1, and the specific formula is as follows:
Figure BDA0003707362800000067
Figure BDA0003707362800000068
in the formula, sgn represents a sign function, w k Deciding hyperplane weights for class k k Is the offset value of the decision function.
In an alternative embodiment of the present invention, the present invention utilizes a traffic status classifier decision weight parameter w in steps S2-2 k ，b k Solving the problem can be equivalent to solving | | w k According to the support vector space constraint, the constraint expression can be set as affine function constraint and further refined as a convex optimization problem under the affine function constraint, and the specific formula is as follows:
Figure BDA0003707362800000071
Figure BDA0003707362800000072
in the formula, epsilon k The allowable range of error for a particular sample point for the class k decision hyperplane, C is the error cost coefficient,
Figure BDA0003707362800000073
is the relaxation variable.
In an optional embodiment of the present invention, the present invention uses the lagrangian multiplier method to solve the step S2-3, and the specific formula is as follows:
Figure BDA0003707362800000074
in the formula of lambda j Is the corresponding lagrange multiplier.
In an optional embodiment of the present invention, in step S2-4, to find the extreme point of the problem, the present invention calculates the partial derivative of each variable to make its value be 0, and the specific formula is as follows:
Figure BDA0003707362800000075
the following can be deduced from the above:
Figure BDA0003707362800000076
s3: solving the classifier decision weight parameters to obtain different traffic states and corresponding vector classifier decision weight parameters;
in an alternative embodiment of the present invention, as shown in fig. 4, step S3 specifically includes the following substeps S3-1 to S3-3:
in an optional embodiment of the present invention, the present invention converts the original problem into the dual problem in steps S3-1, and the specific formula is as follows:
Figure BDA0003707362800000077
in an alternative embodiment of the present invention, step S3-2 of the present invention is to perform an upscaling transformation model by using a gaussian kernel function, and the specific formula is as follows:
Figure BDA0003707362800000081
s.t.λ j ≥0，j＝1，2，...，s
in the formula (I), the compound is shown in the specification,
Figure BDA0003707362800000082
is a kernel function.
In an alternative embodiment of the present invention, the present invention sub-step S3-3 is performed by solving for λ i The optimal solution w can be obtained k * And b k * The concrete formula is as follows:
Figure BDA0003707362800000083
s4: a traffic state classification model based on a self-supervision learning support vector machine is constructed, the traffic state of a highway at a period is estimated by calculating the distance between traffic data of the highway at the period and a traffic state classifier, and the traffic state of the highway at the period is estimated, and the specific formula is as follows:
Figure BDA0003707362800000084
in an alternative embodiment of the invention, as shown in fig. 5, step S4 estimates the effect of the traffic state of the highway during this period.
The method for estimating the traffic state of the expressway based on the self-supervised learning support vector machine of the embodiment utilizes the support vector machine and the self-supervised learning idea to carry out estimation research on the traffic state of the expressway, and fully utilizes a classification model of the support vector machine and a kernel function thereof to reflect the correlation degree of different traffic parameters. The method has better performance for the traffic state estimation of the expressway, and can improve the traffic estimation precision.
The embodiments of the present invention have been described in detail with reference to the drawings, but the present invention is not limited to the above embodiments, and various changes can be made within the knowledge of those skilled in the art without departing from the gist of the present invention.

Claims (6)

1. A highway traffic state estimation method based on an automatic supervision learning support vector machine is characterized by comprising the following steps:
s1: extracting a traffic data characteristic vector of a highway sample, wherein the traffic data characteristic vector of the highway sample comprises flow, interval average speed and occupancy; constructing a traffic state label prejudging device based on self-supervision learning, and performing self-supervision pre-training on traffic state labels on input highway sample traffic data to generate traffic state soft labels; generating a training traffic data set;
s2: constructing a traffic state classification model based on an automatic supervision learning support vector machine, and training the traffic state classification model by using a traffic data characteristic vector of a highway sample;
s3: solving the decision weight parameter of the classifier to obtain different traffic states and the decision weight parameter of the corresponding vector classifier;
s4: and estimating the traffic state of the expressway at the measured time period by calculating the interval distance between the traffic data of the expressway at the measured time period and the traffic state classifier.
2. The method for estimating the traffic state of the expressway based on the automatic supervision learning support vector machine according to claim 1, wherein in the step S1, the process of generating the training traffic data set comprises the following steps:
s11, selecting the flow, the interval average speed and the occupancy as traffic data characteristic vectors of the highway sample, and performing normalization processing, wherein the normalization processing specifically comprises the following steps:
Figure FDA0003707362790000011
in the formula (I), the compound is shown in the specification,
Figure FDA0003707362790000012
the traffic data feature vector is a highway sample traffic data feature vector, comprising three dimensions,
Figure FDA0003707362790000013
average speed of highwayTraffic data, o is the space occupancy of the highway;
s12, constructing a traffic state label prejudge device based on a simple self-training thought in self-supervision learning, and performing self-supervision pre-training on traffic state labels on input highway sample traffic data to generate traffic state soft labels; wherein, the ratio of the average speed to the free flow rate in the selected interval is used as the identification index of the traffic state label prejudger:
Figure FDA0003707362790000014
in the formula, sgn represents a sign function,
Figure FDA0003707362790000015
for the soft tag of the traffic status for the period j,
Figure FDA0003707362790000016
traffic data of average vehicle speed in j time interval v free Is freeway flow rate;
s13, generating a training traffic data set, specifically:
Figure FDA0003707362790000017
wherein T is a training traffic data set,
Figure FDA0003707362790000018
j =1, 2.. And s.s is the total sample number for the characteristic vector of the highway sample traffic data in the j period.
3. The method for estimating traffic state of expressway based on SVM of claim 1, wherein in step S2, the traffic state classification model
Figure FDA0003707362790000021
Comprises the following steps:
Figure FDA0003707362790000022
Figure FDA0003707362790000023
in the formula, sgn represents a sign function, w k Deciding hyperplane weights for class k k For the bias value of the decision function,
Figure FDA0003707362790000024
is a j-period expressway sample traffic data feature vector,
Figure FDA0003707362790000025
for the j-slot traffic status soft tag, j =1, 2.
4. The method for estimating the traffic state of the expressway based on the self-supervised learning support vector machine according to claim 3, wherein in the step S3, solving the decision weight parameters of the classifier to obtain different traffic states and the decision weight parameters of the corresponding vector classifier comprises the following steps:
s31, making a decision weight parameter w of the traffic state classifier k ，b k The solution problem is converted into solution | | | w k And | | l minimization, according to the support vector space constraint, setting the constraint expression of the support vector space constraint as affine function constraint and further refining the affine function constraint as a convex optimization problem under the affine function constraint, wherein the specific formula is as follows:
Figure FDA0003707362790000026
Figure FDA0003707362790000027
in the formula, epsilon k The allowable range of error for a particular sample point for the class k decision hyperplane, C is the error cost coefficient,
Figure FDA0003707362790000028
as a relaxation variable, g (w) k ，b k ) Denotes w k And b k A constraint condition between;
s32, solving the constraint expression by using a Lagrange multiplier method, wherein the specific formula is as follows:
Figure FDA0003707362790000029
in the formula, λ j Is the corresponding lagrange multiplier;
s33, calculating the partial derivatives of the variables to make the values of the variables be 0, wherein the specific formula is as follows:
Figure FDA0003707362790000031
Figure FDA0003707362790000032
Figure FDA0003707362790000033
Figure FDA0003707362790000034
and (3) pushing out:
Figure FDA0003707362790000035
5. the method for estimating the traffic state of the expressway based on the automatic supervision learning support vector machine according to claim 4, wherein in the step S3, the process of solving the decision weight parameters of the classifier to obtain different traffic states and the decision weight parameters of the corresponding vector classifier further comprises the following steps:
s34: converting the solution problem of the constraint expression into a corresponding dual problem, wherein the specific formula is as follows:
Figure FDA0003707362790000036
s.t.λ j ≥0
s35: the Gaussian kernel function is used for carrying out the upscaling conversion model, and the specific formula is as follows:
Figure FDA0003707362790000037
Figure FDA0003707362790000038
s.t.λ j ≥0
in the formula (I), the compound is shown in the specification,
Figure FDA0003707362790000039
is a kernel function; delta is the width of the Gaussian kernel function neuron; q (lambda) j ) Is λ j The solution function of (2); lambda m In order to be a lagrange multiplier,
Figure FDA00037073627900000310
is a soft label of the traffic state in the period of m,
Figure FDA00037073627900000311
at high speed for m time periodA road sample traffic data feature vector, m =1, 2., s, j ≠ m;
s36: by solving for λ i To derive an optimal solution w k * And b k * The concrete formula is as follows:
Figure FDA0003707362790000041
6. the method for estimating traffic state of expressway according to claim 3, wherein in step S4, the following formula is used to estimate the traffic state of the expressway in the measured period by calculating the distance between the traffic data of the measured period and the traffic state classifier:
Figure FDA0003707362790000042
CN202210714620.6A 2022-06-22 2022-06-22 Expressway traffic state estimation method based on self-supervision learning support vector machine Active CN115311844B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202210714620.6A CN115311844B (en) 2022-06-22 2022-06-22 Expressway traffic state estimation method based on self-supervision learning support vector machine

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202210714620.6A CN115311844B (en) 2022-06-22 2022-06-22 Expressway traffic state estimation method based on self-supervision learning support vector machine

Publications (2)

Publication Number Publication Date
CN115311844A true CN115311844A (en) 2022-11-08
CN115311844B CN115311844B (en) 2023-05-16

Family

ID=83855429

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202210714620.6A Active CN115311844B (en) 2022-06-22 2022-06-22 Expressway traffic state estimation method based on self-supervision learning support vector machine

Country Status (1)

Country Link
CN (1) CN115311844B (en)

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107123265A (en) * 2017-06-12 2017-09-01 东南大学 A kind of traffic status of express way method of estimation based on parallel computation
CN112365359A (en) * 2020-11-26 2021-02-12 腾讯科技（深圳）有限公司 Training method, device, equipment and storage medium for planting decision determination model
CN113139651A (en) * 2020-01-20 2021-07-20 北京三星通信技术研究有限公司 Training method and device of label proportion learning model based on self-supervision learning
CN113313145A (en) * 2021-05-10 2021-08-27 东南大学 Expressway traffic incident detection method based on mixed kernel correlation vector machine
CN113781404A (en) * 2021-08-19 2021-12-10 浙江大学 Road disease detection method and system based on self-supervision pre-training
CN114600130A (en) * 2019-12-10 2022-06-07 赫尔实验室有限公司 Process for learning new image classes without labels
EP4012337A1 (en) * 2020-12-11 2022-06-15 Aptiv Technologies Limited Change detection criteria for updating sensor-based reference maps

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107123265A (en) * 2017-06-12 2017-09-01 东南大学 A kind of traffic status of express way method of estimation based on parallel computation
CN114600130A (en) * 2019-12-10 2022-06-07 赫尔实验室有限公司 Process for learning new image classes without labels
CN113139651A (en) * 2020-01-20 2021-07-20 北京三星通信技术研究有限公司 Training method and device of label proportion learning model based on self-supervision learning
CN112365359A (en) * 2020-11-26 2021-02-12 腾讯科技（深圳）有限公司 Training method, device, equipment and storage medium for planting decision determination model
EP4012337A1 (en) * 2020-12-11 2022-06-15 Aptiv Technologies Limited Change detection criteria for updating sensor-based reference maps
CN113313145A (en) * 2021-05-10 2021-08-27 东南大学 Expressway traffic incident detection method based on mixed kernel correlation vector machine
CN113781404A (en) * 2021-08-19 2021-12-10 浙江大学 Road disease detection method and system based on self-supervision pre-training

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
林嘉豪;章宗长;姜冲;郝建业;: "基于生成对抗网络的模仿学习综述" *

Also Published As

Publication number Publication date
CN115311844B (en) 2023-05-16

Similar Documents

Publication Publication Date Title
CN108629073B (en) A kind of degenerative process modeling of multi-mode and method for predicting residual useful life
CN106572493B (en) Rejecting outliers method and system in LTE network
CN108985380B (en) Point switch fault identification method based on cluster integration
Zhou et al. Urban flow prediction with spatial–temporal neural ODEs
CN105069469B (en) A kind of data-flow detection method based on Fuzzy C-Means Cluster Algorithm and entropy theory
Zhao et al. Analysis of users’ electricity consumption behavior based on ensemble clustering
Tian et al. A network traffic prediction method based on IFS algorithm optimised LSSVM
CN109376797B (en) Network traffic classification method based on binary encoder and multi-hash table
Pan et al. UAV anomaly detection using active learning and improved S3VM model
CN116523177A (en) Vehicle energy consumption prediction method and device integrating mechanism and deep learning model
Ren et al. Passenger flow prediction in traffic system based on deep neural networks and transfer learning method
Ye et al. A novel self-supervised learning-based anomalous node detection method based on an autoencoder for wireless sensor networks
Zhang et al. A density-center-based automatic clustering algorithm for IoT data analysis
CN115311844B (en) Expressway traffic state estimation method based on self-supervision learning support vector machine
CN106203526B (en) Goal behavior mode online classification method based on multidimensional characteristic
Huo et al. Traffic anomaly detection method based on improved GRU and EFMS-Kmeans clustering
Wang et al. Contrastive GNN-based traffic anomaly analysis against imbalanced dataset in IoT-based its
Dan et al. A dynamic model for urban population density estimation using mobile phone location data
CN109859467A (en) A kind of mining analysis method of Environmental Factors in traffic model
Mao et al. Naive Bayesian algorithm classification model with local attribute weighted based on KNN
Cui et al. Spatio-temporal broad learning networks for traffic speed prediction
Liu et al. Uncovering the aggregation pattern of gps trajectory based on spatiotemporal clustering and 3D visualization
CN116189026A (en) Pedestrian re-recognition method and device and storage medium
Ghaedi et al. Improving Electricity Theft Detection using Combination of Improved Crow Search Algorithm and Support Vector Machine
Wendl et al. Novelty detection in very high resolution urban scenes with density forests

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