CN115457764B - Road section traffic density estimation method, device and medium based on vehicle track data - Google Patents
Road section traffic density estimation method, device and medium based on vehicle track data Download PDFInfo
- Publication number
- CN115457764B CN115457764B CN202211018930.0A CN202211018930A CN115457764B CN 115457764 B CN115457764 B CN 115457764B CN 202211018930 A CN202211018930 A CN 202211018930A CN 115457764 B CN115457764 B CN 115457764B
- Authority
- CN
- China
- Prior art keywords
- traffic density
- neural network
- time slice
- value
- track data
- 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
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
- G08G1/0108—Measuring and analyzing of parameters relative to traffic conditions based on the source of data
-
- 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
- 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
- G08G1/0145—Measuring and analyzing of parameters relative to traffic conditions for specific applications for active traffic flow control
-
- 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
Abstract
The invention discloses a road section traffic density estimation method, a device and a medium based on vehicle track data, wherein the method comprises the following steps: acquiring vehicle track data and preprocessing the vehicle track data; acquiring an estimated time interval, and slicing vehicle track data; extracting characteristics of vehicle track data, constructing an original characteristic matrix of each time slice, and carrying out normalization processing to obtain an input characteristic matrix; inputting the input feature matrix into a plurality of preset neural network models, and training and outputting the original traffic density estimated value of each neural network model for each time slice; correcting the estimation result of the traffic density of each time slice; and fusing a plurality of neural network model estimation results. The invention fully excavates the characteristic value of the vehicle track data, fuses a plurality of intelligent algorithms, can realize more accurate road section traffic density estimation, and provides important parameters for traffic control and management. The invention can be widely applied to the field of intelligent traffic information acquisition.
Description
Technical Field
The invention relates to the field of intelligent traffic information acquisition, in particular to a road section traffic density estimation method, device and medium based on vehicle track data.
Background
Traffic density is an important parameter in the traffic field, and can be widely applied to traffic control and management. By combining with the signal timing scheme, the traffic density has wide application prospect in expressway ramp control and urban road active control. Traffic jam problems exist in the current expressways and urban internal arterial roads in specific peak hours. Some navigation companies provide intelligent traffic control schemes based on indexes such as queuing length, but the method has poor control effect on road sections without queuing; the ramp control, active control and other modes are basically the total number of vehicles in the control road section, namely the traffic density is controlled, so that the control effect on various road traffic states is good.
The existing traffic density acquisition technology is mainly divided into a fixed type and a movable type: the fixed acquisition mode is often based on hardware facilities, has the problems of easy environmental influence, visual field blind area, low precision, high cost, inconvenient maintenance and the like, and has unsustainability in the method because a helicopter is needed to be used for acquisition; the mobile acquisition mode is mostly based on floating car data, namely part of car track data collected by a vehicle-mounted GPS. The increasing popularity of navigation software provides convenience for the collection of floating car datasets. The mobile acquisition cost is low, the applicability is wide, the advantages are obvious, and the research of estimating the traffic density by utilizing the track data of the floating car has higher application value.
Currently, in the field of intelligent traffic information collection based on floating car data, most of main students use a machine learning model in the field of artificial intelligence to study traffic estimation. However, the traffic density is taken as an important parameter for traffic management and control, the related research results are few, and the research results at home and abroad have certain defects, for example, the characteristic value extraction is only limited to a macroscopic-macroscopic layer, the characteristic value traffic characteristics are not fully mined, and the like. Even though models proposed by some scholars have a certain competitiveness in accuracy, the high requirement on the quality of the data set is still difficult to adapt to the current situation of urban traffic density acquisition.
Disclosure of Invention
In order to solve at least one of the technical problems existing in the prior art to a certain extent, the invention aims to provide a road traffic density estimation method, device and medium based on vehicle track data.
The technical scheme adopted by the invention is as follows:
a road section traffic density estimation method based on vehicle track data comprises the following steps:
acquiring vehicle track data and preprocessing the vehicle track data;
acquiring an estimated time interval, and slicing vehicle track data according to the estimated time interval;
extracting characteristics of vehicle track data, constructing an original characteristic matrix of each time slice, and carrying out normalization processing to obtain an input characteristic matrix;
inputting the input feature matrix into a plurality of preset neural network models, and training and outputting the original traffic density estimated value of each neural network model for each time slice by combining a sliding time window principle;
correcting the estimation result of the traffic density of each time slice to obtain a corrected traffic density estimation value;
and fusing a plurality of neural network model estimation results to improve the accuracy of the traffic density estimation value of each time slice.
Further, the vehicle track data comprises a vehicle number, a recording time, relative coordinates, a vehicle speed and an acceleration;
the preprocessing of the vehicle track data includes:
acquiring a first vehicle corresponding moment, an X coordinate and a Y coordinate recorded by a data set as reference values;
and carrying out difference between the recording time, the X coordinate and the Y coordinate in the rest vehicle track data and the reference value to obtain the preprocessed vehicle track data.
Further, the slicing the vehicle track data according to the estimated time interval includes:
slicing the vehicle track data according to the estimated time intervals, marking each time slice number as i, and further dividing the time slices into n time intervals;
the features include distance features, motion features, contrast features and traffic features, and the input feature matrix M' i of the time slice i is constructed by n，m ：
Constructing an original feature matrix Mi of a time slice i n，m The method comprises the following steps:
wherein the jth row and the kth column of the element P j，k Indicating the j (j=1, 2., n.) th (k=1, 2., m.) original feature values in the j (j=1, 2., n.) th time interval; extracting m characteristic values from each time interval j;
the original characteristic value P is processed j，k Normalizing to obtain an input feature matrix M' i of the time slice i n，m 。
Further, the input feature matrix M' i n，m The expression of (2) is:
wherein, the liquid crystal display device comprises a liquid crystal display device,representing the average value of the kth characteristic element of n time intervals in the time slice i, i.e. the original characteristic matrix Mi n，m The k-th column average value of (b); s is(s) k Representing standard deviation of kth characteristic element of n time intervals in time slice i, namely original characteristic matrix Mi n，m The kth column standard deviation; p'. j，k For the original characteristic value P j，k Normalized values, representing the input feature values of the k (k=1, 2,) th (j=1, 2.., m) th time interval; m' i n，m Representing the input feature matrix of time slice i.
Further, the m eigenvalues extracted from each time interval j include:
the feature value of the distance feature comprises: average value of coordinate point distance of adjacent vehiclesStandard deviation s of coordinate point distance of adjacent vehicles h Distance range R of adjacent vehicle coordinate points h Neighboring vehicle coordinate point distance square average +.>Adjacent vehicle coordinate point distance square standard deviation +.>Distance square range of adjacent vehicle coordinate points +.>
The characteristic value of the motion characteristic comprises: average value of speedStandard deviation of speed s v Extremely poor speed R v Acceleration average->Standard deviation of acceleration s a Acceleration range R a ；
The characteristic value of the contrast characteristic comprises: the difference between the speed averages of the j+1 and j time intervalsDifference of speed standard deviation->Difference in speed difference->Difference between average acceleration values->Difference of standard deviation of acceleration->Difference of acceleration difference->
The characteristic value of the traffic characteristic comprises: lane number W, traffic volume V;
the traffic volume V is calculated in the following way:
wherein t is 0 Representing the average value of the length L and the speed of the road sectionT represents the ratio of the road length L to the maximum speed v max Ratio of; a. beta and c are constants.
Further, the neural network model comprises a long-term and short-term memory neural network model, an extreme gradient lifting neural network model, a multi-layer perceptron neural network model and a random forest neural network model;
the method for inputting the input feature matrix into a plurality of preset neural network models and training and outputting the estimated value of the original traffic density of each neural network model for each time slice by combining the sliding time window principle comprises the following steps:
input feature matrix M' i of time slice i n，m Inputting a p-th neural network model, and acquiring a more accurate estimated value by utilizing the smoothness of a time sequence by combining a sliding time window principle;
training and outputting, namely obtaining traffic density estimated values of the time slice i and the time slice i+1 under the input of the input feature matrix;
discarding the traffic density estimated value of the time slice i+1, and taking the traffic density estimated value d of the time slice i p，i As an original traffic density estimate of the input feature matrix.
Further, the correcting the estimation result of the traffic density of each time slice to obtain a corrected traffic density estimation value includes:
correcting the original traffic density estimated value d of the time slice i output by the p-th neural network model by filtering p，t To obtain a corrected density estimation value d 'with smaller error than the true value' p，t The calculation formula is as follows:
d′ p，i ＝f(d p，i-2 ，d p，i-1 ，d p，i )
wherein d i-2 ，d i-1 ，d i The original traffic density estimated values of the time slices i-2, i-1 and i are respectively represented; f () is a mapping function.
Further, the fusing the plurality of neural network model estimation results includes:
the Bayesian algorithm is used for fusing a plurality of neural network model estimation results, and the calculation formula is as follows:
D i ＝g(d′ 1，i ，......，d′ q，i )
wherein d' p，i Representing corrected time slice i density estimates obtained using a p-th neural network model, p=1, 2. D (D) i Representing the density estimated value of the fused time slice i; the function g represents the operation of the bayesian neural network.
The invention adopts another technical scheme that:
a road segment traffic density estimation device based on vehicle trajectory data, comprising:
at least one processor;
at least one memory for storing at least one program;
the at least one program, when executed by the at least one processor, causes the at least one processor to implement the method described above.
The invention adopts another technical scheme that:
a computer readable storage medium, in which a processor executable program is stored, which when executed by a processor is adapted to carry out the method as described above.
The beneficial effects of the invention are as follows: the invention fully excavates the characteristic value of the vehicle track data, fuses a plurality of intelligent algorithms, can realize more accurate road section traffic density estimation, and provides important parameters for traffic control and management.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following description is made with reference to the accompanying drawings of the embodiments of the present invention or the related technical solutions in the prior art, and it should be understood that the drawings in the following description are only for convenience and clarity of describing some embodiments in the technical solutions of the present invention, and other drawings may be obtained according to these drawings without the need of inventive labor for those skilled in the art.
Fig. 1 is a flow chart of a road traffic density estimation method based on vehicle track data according to an embodiment of the invention;
FIG. 2 is a schematic diagram of a concept of estimating traffic density of a road section using vehicle trajectory data according to an embodiment of the present invention;
FIG. 3 is a schematic diagram of the LSTM neural network model input and output in an embodiment of the invention;
FIG. 4 is a schematic diagram of filtering correction of an original traffic density estimation value according to an embodiment of the present invention;
fig. 5 is a schematic diagram of a result of estimating a road segment density of a state dataset according to an embodiment of the present invention.
Detailed Description
Embodiments of the present invention are described in detail below, examples of which are illustrated in the accompanying drawings, wherein like or similar reference numerals refer to like or similar elements or elements having like or similar functions throughout. The embodiments described below by referring to the drawings are illustrative only and are not to be construed as limiting the invention. The step numbers in the following embodiments are set for convenience of illustration only, and the order between the steps is not limited in any way, and the execution order of the steps in the embodiments may be adaptively adjusted according to the understanding of those skilled in the art.
In the description of the present invention, it should be understood that references to orientation descriptions such as upper, lower, front, rear, left, right, etc. are based on the orientation or positional relationship shown in the drawings, are merely for convenience of description of the present invention and to simplify the description, and do not indicate or imply that the apparatus or elements referred to must have a particular orientation, be constructed and operated in a particular orientation, and thus should not be construed as limiting the present invention.
In the description of the present invention, a number means one or more, a number means two or more, and greater than, less than, exceeding, etc. are understood to not include the present number, and above, below, within, etc. are understood to include the present number. The description of the first and second is for the purpose of distinguishing between technical features only and should not be construed as indicating or implying relative importance or implicitly indicating the number of technical features indicated or implicitly indicating the precedence of the technical features indicated.
In the description of the present invention, unless explicitly defined otherwise, terms such as arrangement, installation, connection, etc. should be construed broadly and the specific meaning of the terms in the present invention can be reasonably determined by a person skilled in the art in combination with the specific contents of the technical scheme.
As shown in fig. 1, the present embodiment provides a road traffic density estimation method based on vehicle track data, which fully exerts advantages of a floating vehicle data set, and fuses various intelligent algorithms to realize more accurate road traffic density estimation, thereby providing important parameters for traffic control and management. The method specifically comprises the following steps:
s1, acquiring vehicle track data and preprocessing the vehicle track data.
The vehicle track data comprises a vehicle number, a recording time, relative coordinates, a vehicle speed and an acceleration, and the data preprocessing comprises the following steps:
designating the corresponding moment, X coordinate and Y coordinate of the first vehicle recorded by the data set as reference values;
and the recording time, the X coordinate and the Y coordinate in the rest vehicle track data are differenced from the reference value, so as to obtain the track data after preprocessing.
As an alternative embodiment, step S1 includes the following S11-S12:
s11, acquiring a data set for estimating the road traffic density, wherein parameters such as vehicle numbers, recording moments, relative coordinates, vehicle speed and acceleration are needed to be included. In the embodiment, a center urban floating car data set of 5 months and 5 days in 2018 of Fuzhou is selected, 6,402,027 GPS sample points of 7073 taxis in the center area of Fuzhou within 1 day are included, the data interval is unequal (5-20 seconds), the density is high, and the tide phenomenon is obvious.
As shown in fig. 2, the idea of the present invention is to estimate the overall traffic density of a road section by means of the track data uploaded by a part of vehicles. In this embodiment, a vehicle with a random extraction ratio of 15% is used as a part of vehicles for estimation, the trajectory data thereof is used for estimating traffic density, and then the estimated value is compared with the true value of the whole sample to obtain the estimation accuracy.
S12, preprocessing track data of the part of vehicles, wherein the method comprises the following steps of:
designating the corresponding moment, X coordinate and Y coordinate of the first vehicle recorded by the data set as reference values;
the recording time, the X coordinate, the Y coordinate and the reference value are differenced from the rest of the vehicle track data to obtain the track data after preprocessing, as shown in table 1.
TABLE 1 partial track data after pretreatment
S2, acquiring an estimated time interval, and slicing the vehicle track data according to the estimated time interval.
Selecting an estimated time interval, slicing the overall data according to the estimated time interval, and further dividing the time slice into n time intervals.
As an alternative embodiment, step S2 includes the following contents S21-S23:
s21, comprehensively considering the sample interval condition of the data set and the traffic density estimation requirement, and selecting an estimation time interval to be 3 minutes.
S22, slicing the overall data according to the estimated time interval, and marking the serial number of each time slice as i.
S23, further dividing the time slice into n time intervals, n=6, i.e. 30 seconds for each time interval.
And S3, extracting characteristics of vehicle track data, constructing an original characteristic matrix of each time slice, and carrying out normalization processing to obtain an input characteristic matrix.
Extracting characteristic values contained in the track data, constructing an original characteristic matrix of each time slice, and carrying out normalization processing to obtain an input characteristic matrix. Wherein the characteristic values are classified into distance characteristics, motion characteristics, contrast characteristics, traffic characteristics and the like.
As an alternative embodiment, step S3 includes the following S31-S33:
s31, extracting original characteristic values contained in the track data, wherein the original characteristic values can be divided into distance characteristics, motion characteristics, contrast characteristics, traffic characteristics and the like.
The distance characteristic comprises a distance average value of coordinate points of adjacent vehiclesStandard deviation s of coordinate point distance of adjacent vehicles h Distance range R of adjacent vehicle coordinate points h Neighboring vehicle coordinate point distance square average +.>Adjacent vehicle coordinate point distance square standard deviation +.>Distance square range of adjacent vehicle coordinate points +.>
The motion characteristics include a velocity averageStandard deviation of speed s v Extremely poor speed R v Acceleration average->Standard deviation of acceleration s a Acceleration range R a ；
The contrast characteristic comprises the difference between the j+1 and the j time interval velocity average valueDifference of speed standard deviation->Difference in speed difference->Difference between average acceleration values->Difference of standard deviation of accelerationDifference of acceleration difference->
The traffic characteristics include lane number W, traffic volume V. The traffic volume is obtained through calculation of a path resistance function variation, and the formula is as follows:
wherein t is 0 Representing the average value of the length L and the speed of the road sectionT represents the ratio of the road length L to the maximum speed v max Ratio of; the constant a and the constant beta respectively take default values of 0.15 and 0.4; c is replaced by a constant based on the actual situation of the road section and the history data.
S32, constructing an input feature matrix M' i of the time slice i n，m ：
Wherein the jth row and the kth column of the element P j，k The j (j=1, 2.) th time interval k (k=1, 2.) th, m.) th raw feature values are indicated.
S33, carrying out normalization processing on the original feature matrix. The original characteristic value P is processed j，k Put into formulaColumn-by-column normalization to obtain an input feature matrix M' i of the time slice i n，m The calculation formula is as follows:
wherein, the liquid crystal display device comprises a liquid crystal display device,representing the average value of the kth characteristic element of n time intervals in the time slice i, i.e. the original characteristic matrix Mi n，m The k-th column average value of (b); s is(s) k Representing standard deviation of kth characteristic element of n time intervals in time slice i, namely original characteristic matrix Mi n，m The kth column standard deviation; p'. j，k For the original characteristic value P j，k Normalized values, representing the input feature values of the k (k=1, 2,) th (j=1, 2.., m) th time interval; m' i n，m Representing the input feature matrix of time slice i.
S4, inputting the input feature matrix into a plurality of preset neural network models, and training and outputting the original traffic density estimated value of each neural network model for each time slice.
And inputting the feature matrix into a neural network model, and training and outputting the original traffic density estimated value of each neural network model for each time slice by combining a sliding time window principle. The neural network model comprises long-term memory (LSTM), extreme gradient lifting (XGBOOST), multi-layer perceptron (MLP), random Forest (RF) and the like.
Further as an alternative embodiment, the input feature matrix M' i of the time slice i is n，m The p (p=1, 2,3, 4) neural network model is input, and in combination with the sliding time window principle, in order to obtain more accurate estimated values by utilizing the smoothness of the time sequence, the traffic density estimated values of the time slices i and i+1 obtained under the input of the feature matrix are trained and output. Discarding the estimated value of the time slice i+1, and taking the estimated value d of the time slice i p，i Is saidThe original estimate of the feature matrix. Wherein the neural network model comprises LSTM, XGBOOST, MLP, RF and the like. As shown in fig. 3, taking LSTM as an example, the neural network model input-output flow is shown.
S5, correcting the estimation result of the traffic density of each time slice to obtain a corrected traffic density estimation value.
And correcting the estimation result of the traffic density of each time slice by using filtering to obtain a corrected traffic density estimation value so as to reduce the error between the estimation value and the true value.
Further alternatively, filtering is used to correct the original traffic density estimated value d of the time slice i output by the p-th neural network model p，i To obtain a corrected density estimation value d 'with smaller error than the true value' p，t The calculation formula is as follows:
d′ p，i ＝f(d p，i-2 ，d p，i-1 ，d p，i )
wherein d i-2 ，d i-1 ，d i The original traffic density estimated values of the time slices i-2, i-1 and i are respectively represented;
as shown in fig. 4, the mapping function f is determined according to the idea of minimizing the difference between the density estimation value and the true value in the training data set of the road segment.
S6, fusing a plurality of neural network model estimation results to improve the accuracy of the traffic density estimation value of each time slice.
And fusing a plurality of neural network model estimation results by using a Bayesian algorithm so as to improve the accuracy of the traffic density estimation value of each time slice.
Further as an optional implementation manner, a bayesian algorithm is used to fuse a plurality of neural network model estimation results, and a calculation formula is as follows:
D i ＝g(d′ 1，i ，......，d′ q，i )
wherein d' p，i Representing a corrected time slice i density estimated value obtained by using a p-th neural network model; p=1, 2.,. Q; d (D) i Representing the density estimated value of the fused time slice i; function g represents shellfishThe operation of the phyllus neural network is as follows:
wherein P (A) is the prior probability and P (B) is evidence;
the mean square error and the accuracy of the estimated value obtained by each method are shown in table 2, and the accuracy of the estimated value obtained after Bayesian fusion is the highest; wherein the unit of the mean square error is the square of the number of vehicles per kilometer, i.e. (pcu/km) 2 。
Table 2 mean square error and accuracy of the estimates obtained by the methods
In the embodiment of the invention, the estimated result of the road section density of the Fuzhou dataset is shown in fig. 5.
In summary, compared with the prior art, the embodiment has the following advantages and beneficial effects:
(1) The invention provides a new idea of estimating the traffic density of the road section based on partial vehicle track data, and based on fully excavating microscopic characteristics and traffic characteristics of floating vehicle data, a plurality of machine learning algorithms are fused to obtain a more accurate traffic density estimated value of the road section. The intelligent traffic information acquisition method designed by the invention is based on the combined driving of data-model, has important value for traffic control and management, and can be widely applied to scenes such as ramp control, active control, congestion judgment and the like.
(2) The invention fully excavates the data characteristics of the floating car, and improves the robustness of traffic density estimation and the adaptability to different quality data sets. When an input matrix is constructed, the characteristic values are divided into categories such as distance characteristics, motion characteristics, contrast characteristics, traffic characteristics and the like, characteristics such as coordinate point distances and the like are extracted on a microscopic level, traffic characteristics such as a road resistance function and the like are creatively introduced, and the connection between density estimation and traffic discipline knowledge is enhanced.
(3) The invention adopts various methods to improve the estimation precision, not only introduces a sliding time window and filters to correct the estimation value, but also utilizes a Bayesian algorithm to fuse a plurality of neural network model estimation results. The method can reduce the error between the estimated value and the true value and improve the accuracy of traffic density estimation.
The embodiment also provides a road traffic density estimating device based on vehicle track data, which comprises:
at least one processor;
at least one memory for storing at least one program;
the at least one program, when executed by the at least one processor, causes the at least one processor to implement the method illustrated in fig. 1.
The road section traffic density estimation device based on the vehicle track data can execute the road section traffic density estimation method based on the vehicle track data, can execute any combination implementation steps of the method embodiments, and has the corresponding functions and beneficial effects of the method.
The present application also discloses a computer program product or a computer program comprising computer instructions stored in a computer readable storage medium. The computer instructions may be read from a computer-readable storage medium by a processor of a computer device, and executed by the processor, to cause the computer device to perform the method shown in fig. 1.
The embodiment also provides a storage medium which stores instructions or programs for executing the road section traffic density estimation method based on the vehicle track data, and when the instructions or programs are run, the instructions or programs can execute the steps in any combination of the embodiments of the method, and the method has the corresponding functions and beneficial effects.
In some alternative embodiments, the functions/acts noted in the block diagrams may occur out of the order noted in the operational illustrations. For example, two blocks shown in succession may in fact be executed substantially concurrently or the blocks may sometimes be executed in the reverse order, depending upon the functionality/acts involved. Furthermore, the embodiments presented and described in the flowcharts of the present invention are provided by way of example in order to provide a more thorough understanding of the technology. The disclosed methods are not limited to the operations and logic flows presented herein. Alternative embodiments are contemplated in which the order of various operations is changed, and in which sub-operations described as part of a larger operation are performed independently.
Furthermore, while the invention is described in the context of functional modules, it should be appreciated that, unless otherwise indicated, one or more of the described functions and/or features may be integrated in a single physical device and/or software module or one or more functions and/or features may be implemented in separate physical devices or software modules. It will also be appreciated that a detailed discussion of the actual implementation of each module is not necessary to an understanding of the present invention. Rather, the actual implementation of the various functional modules in the apparatus disclosed herein will be apparent to those skilled in the art from consideration of their attributes, functions and internal relationships. Accordingly, one of ordinary skill in the art can implement the invention as set forth in the claims without undue experimentation. It is also to be understood that the specific concepts disclosed are merely illustrative and are not intended to be limiting upon the scope of the invention, which is to be defined in the appended claims and their full scope of equivalents.
The functions, if implemented in the form of software functional units and sold or used as a stand-alone product, may be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention may be embodied essentially or in a part contributing to the prior art or in a part of the technical solution, in the form of a software product stored in a storage medium, comprising several instructions for causing a computer device (which may be a personal computer, a server, a network device, etc.) to perform all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a random access Memory (RAM, random Access Memory), a magnetic disk, or an optical disk, or other various media capable of storing program codes.
Logic and/or steps represented in the flowcharts or otherwise described herein, e.g., a ordered listing of executable instructions for implementing logical functions, can be embodied in any computer-readable medium for use by or in connection with an instruction execution system, apparatus, or device, such as a computer-based system, processor-containing system, or other system that can fetch the instructions from the instruction execution system, apparatus, or device and execute the instructions. For the purposes of this description, a "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transport the program for use by or in connection with the instruction execution system, apparatus, or device.
More specific examples (a non-exhaustive list) of the computer-readable medium would include the following: an electrical connection (electronic device) having one or more wires, a portable computer diskette (magnetic device), a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disc read-only memory (CDROM). In addition, the computer readable medium may even be paper or other suitable medium on which the program is printed, as the program may be electronically captured, via, for instance, optical scanning of the paper or other medium, then compiled, interpreted or otherwise processed in a suitable manner, if necessary, and then stored in a computer memory.
It is to be understood that portions of the present invention may be implemented in hardware, software, firmware, or a combination thereof. In the above-described embodiments, the various steps or methods may be implemented in software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, may be implemented using any one or combination of the following techniques, as is well known in the art: discrete logic circuits having logic gates for implementing logic functions on data signals, application specific integrated circuits having suitable combinational logic gates, programmable Gate Arrays (PGAs), field Programmable Gate Arrays (FPGAs), and the like.
In the foregoing description of the present specification, reference has been made to the terms "one embodiment/example", "another embodiment/example", "certain embodiments/examples", and the like, means that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the invention. In this specification, schematic representations of the above terms do not necessarily refer to the same embodiments or examples. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.
While embodiments of the present invention have been shown and described, it will be understood by those of ordinary skill in the art that: many changes, modifications, substitutions and variations may be made to the embodiments without departing from the spirit and principles of the invention, the scope of which is defined by the claims and their equivalents.
While the preferred embodiment of the present invention has been described in detail, the present invention is not limited to the above embodiments, and various equivalent modifications and substitutions can be made by those skilled in the art without departing from the spirit of the present invention, and these equivalent modifications and substitutions are intended to be included in the scope of the present invention as defined in the appended claims.
Claims (6)
1. The road section traffic density estimation method based on the vehicle track data is characterized by comprising the following steps of:
acquiring vehicle track data and preprocessing the vehicle track data;
acquiring an estimated time interval, and slicing vehicle track data according to the estimated time interval;
extracting characteristics of vehicle track data, constructing an original characteristic matrix of each time slice, and carrying out normalization processing to obtain an input characteristic matrix;
inputting the input feature matrix into a plurality of preset neural network models, and training and outputting the original traffic density estimated value of each neural network model for each time slice;
correcting the estimation result of the traffic density of each time slice to obtain a corrected traffic density estimation value;
fusing a plurality of neural network model estimation results to improve the accuracy of each time slice traffic density estimation value;
the slicing the vehicle track data according to the estimated time interval includes:
slicing the vehicle track data according to the estimated time intervals, marking each time slice number as i, and further dividing the time slices into n time intervals;
the features include distance features, motion features, contrast features and traffic features, and the input feature matrix M' i of the time slice i is constructed by n,m ：
Constructing an original feature matrix Mi of a time slice i n,m The method comprises the following steps:
wherein the jth row and the kth column of the element P j,k Representing a kth original characteristic value in a jth time interval; extracting m characteristic values from each time interval j; j=1, 2..the term "n, k=1, 2..the term" m;
the original characteristic value P is processed j,k Normalizing to obtain an input feature matrix M' i of the time slice i n,m ；
The input feature matrix M' i n,m The expression of (2) is:
wherein, the liquid crystal display device comprises a liquid crystal display device,representing the average value of the kth characteristic element of n time intervals in the time slice i, i.e. the original characteristic matrix Mi n,m The k-th column average value of (b); s is(s) k Representing standard deviation of kth characteristic element of n time intervals in time slice i, namely original characteristic matrix Mi n,m The kth column standard deviation; p'. j,k For the original characteristic value P j,k Normalized values represent the kth input characteristic value in the jth time interval; m' i n,m An input feature matrix representing a time slice i;
the m eigenvalues extracted from each time interval j include:
the feature value of the distance feature comprises: average value of coordinate point distance of adjacent vehiclesStandard deviation s of coordinate point distance of adjacent vehicles h Distance range R of adjacent vehicle coordinate points h Neighboring vehicle coordinate point distance square average +.>Adjacent vehicle coordinate point distance square standard deviation s h2 Distance square polar difference R between adjacent vehicle coordinate points h2 ；
The characteristic value of the motion characteristic comprises: average value of speedStandard deviation of speed s v Extremely poor speed R v Average value of accelerationStandard deviation of acceleration s a Acceleration range R a ；
The characteristic value of the contrast characteristic comprises: the difference between the speed averages of the j+1 and j time intervalsDifference of speed standard deviation->Difference in speed difference->Difference between average acceleration values->Difference of standard deviation of accelerationDifference of acceleration difference->
The characteristic value of the traffic characteristic comprises: lane number W, traffic volume V;
the traffic volume V is calculated in the following way:
wherein t is 0 Representing the average value of the length L and the speed of the road sectionT represents the ratio of the road length L to the maximum speed v max Ratio of; a. beta, c are constants;
the neural network model comprises a long-term and short-term memory neural network model, an extreme gradient lifting neural network model, a multi-layer perceptron neural network model and a random forest neural network model;
the input feature matrix is input into a plurality of preset neural network models, and the training output of the original traffic density estimated value of each neural network model for each time slice comprises the following steps:
input feature matrix M' i of time slice i n,m Inputting a p-th neural network model, and acquiring a more accurate estimated value by utilizing the smoothness of a time sequence by combining a sliding time window principle;
training and outputting, namely obtaining traffic density estimated values of the time slice i and the time slice i+1 under the input of the input feature matrix;
discarding the traffic density estimated value of the time slice i+1, and taking the traffic density estimated value d of the time slice i p,i As an original traffic density estimate of the input feature matrix.
2. The method for estimating traffic density on a road segment based on vehicle trajectory data according to claim 1, wherein the vehicle trajectory data includes a vehicle number, a recording time, a relative coordinate, a vehicle speed and an acceleration;
the preprocessing of the vehicle track data includes:
acquiring a first vehicle corresponding moment, an X coordinate and a Y coordinate recorded by a data set as reference values;
and carrying out difference between the recording time, the X coordinate and the Y coordinate in the rest vehicle track data and the reference value to obtain the preprocessed vehicle track data.
3. The method for estimating traffic density on a road segment based on vehicle trajectory data according to claim 1, wherein said correcting the estimated result of traffic density of each time slice to obtain a corrected estimated value of traffic density comprises:
correcting the original traffic density estimated value d of the time slice i output by the p-th neural network model by filtering p,i To obtain a corrected density estimation value d 'with smaller error than the true value' p,i The calculation formula is as follows:
d′ p,i ＝f(d p,i-2 ,d p,i-1 ,d p,i )
wherein d p,i-2 ,d p,i-1 ,d p,i The original traffic density estimated values of the time slices i-2, i-1 and i are respectively represented; f () is a mapping function.
4. The method for estimating traffic density on a road segment based on vehicle trajectory data according to claim 1, wherein the fusing the plurality of neural network model estimation results includes:
the Bayesian algorithm is used for fusing a plurality of neural network model estimation results, and the calculation formula is as follows:
D i ＝g(d′ 1,i ,……,d′ q,i )
wherein d' p,i Representing a corrected time slice i density estimate obtained using a p-th neural network model, p=1, 2 … …, q; d (D) i Representing the density estimated value of the fused time slice i; the function g represents the operation of the bayesian neural network.
5. A road traffic density estimation device based on vehicle trajectory data, comprising:
at least one processor;
at least one memory for storing at least one program;
the at least one program, when executed by the at least one processor, causes the at least one processor to implement the method of any one of claims 1-4.
6. A computer readable storage medium, in which a processor executable program is stored, characterized in that the processor executable program is for performing the method according to any of claims 1-4 when being executed by a processor.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211018930.0A CN115457764B (en) | 2022-08-24 | 2022-08-24 | Road section traffic density estimation method, device and medium based on vehicle track data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211018930.0A CN115457764B (en) | 2022-08-24 | 2022-08-24 | Road section traffic density estimation method, device and medium based on vehicle track data |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115457764A CN115457764A (en) | 2022-12-09 |
| CN115457764B true CN115457764B (en) | 2023-07-18 |
Family
ID=84299437
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211018930.0A Active CN115457764B (en) | 2022-08-24 | 2022-08-24 | Road section traffic density estimation method, device and medium based on vehicle track data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115457764B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117152973B (en) * | 2023-10-27 | 2024-01-05 | 贵州宏信达高新科技有限责任公司 | Expressway real-time flow monitoring method and system based on ETC portal data |
| CN117253364B (en) * | 2023-11-15 | 2024-01-26 | 南京感动科技有限公司 | Traffic jam event extraction and situation fusion method and system |
Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2001084479A (en) * | 1999-09-14 | 2001-03-30 | Matsushita Electric Ind Co Ltd | Method and device for forecasting traffic flow data |
| JP2013210941A (en) * | 2012-03-30 | 2013-10-10 | I-Transport Lab Co Ltd | Traffic flow prediction device, traffic flow prediction method, and traffic flow prediction program |
| JP2013214232A (en) * | 2012-04-03 | 2013-10-17 | Sumitomo Electric Ind Ltd | Traffic information prediction device, method for predicting traffic information, and computer program |
| KR101530636B1 (en) * | 2014-11-11 | 2015-06-23 | 한국건설기술연구원 | Apparatus and Method for obtaining traffic density by counting cars number in unit section |
| WO2016096226A1 (en) * | 2014-12-18 | 2016-06-23 | Be-Mobile Nv | A traffic data fusion system and the related method for providing a traffic state for a network of roads |
| CN109272746A (en) * | 2018-08-20 | 2019-01-25 | 广东交通职业技术学院 | A kind of MFD estimating and measuring method based on BP neural network data fusion |
| CN109754605A (en) * | 2019-02-27 | 2019-05-14 | 中南大学 | A kind of traffic forecast method based on attention temporal diagram convolutional network |
| CN111341099A (en) * | 2020-02-27 | 2020-06-26 | 阿里巴巴集团控股有限公司 | Data processing method and device and electronic equipment |
| KR102245580B1 (en) * | 2020-09-22 | 2021-04-29 | 재단법인차세대융합기술연구원 | Control server for estimating traffic density using adas probe data |
| CN112884222A (en) * | 2021-02-10 | 2021-06-01 | 武汉大学 | Time-period-oriented LSTM traffic flow density prediction method |
| CN114333335A (en) * | 2022-03-15 | 2022-04-12 | 成都交大大数据科技有限公司 | Lane-level traffic state estimation method, device and system based on track data |
| WO2022077767A1 (en) * | 2020-10-16 | 2022-04-21 | 深圳先进技术研究院 | Traffic flow prediction method and apparatus, computer device, and readable storage medium |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| GB201802366D0 (en) * | 2018-02-14 | 2018-03-28 | Tom Tom Traffic B V | Methods and systems for generating taffic volumn or traffic density data |
| WO2020027864A1 (en) * | 2018-07-31 | 2020-02-06 | Didi Research America, Llc | System and method for point-to-point traffic prediction |
-
2022
- 2022-08-24 CN CN202211018930.0A patent/CN115457764B/en active Active
Patent Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2001084479A (en) * | 1999-09-14 | 2001-03-30 | Matsushita Electric Ind Co Ltd | Method and device for forecasting traffic flow data |
| JP2013210941A (en) * | 2012-03-30 | 2013-10-10 | I-Transport Lab Co Ltd | Traffic flow prediction device, traffic flow prediction method, and traffic flow prediction program |
| JP2013214232A (en) * | 2012-04-03 | 2013-10-17 | Sumitomo Electric Ind Ltd | Traffic information prediction device, method for predicting traffic information, and computer program |
| KR101530636B1 (en) * | 2014-11-11 | 2015-06-23 | 한국건설기술연구원 | Apparatus and Method for obtaining traffic density by counting cars number in unit section |
| WO2016096226A1 (en) * | 2014-12-18 | 2016-06-23 | Be-Mobile Nv | A traffic data fusion system and the related method for providing a traffic state for a network of roads |
| CN109272746A (en) * | 2018-08-20 | 2019-01-25 | 广东交通职业技术学院 | A kind of MFD estimating and measuring method based on BP neural network data fusion |
| CN109754605A (en) * | 2019-02-27 | 2019-05-14 | 中南大学 | A kind of traffic forecast method based on attention temporal diagram convolutional network |
| CN111341099A (en) * | 2020-02-27 | 2020-06-26 | 阿里巴巴集团控股有限公司 | Data processing method and device and electronic equipment |
| KR102245580B1 (en) * | 2020-09-22 | 2021-04-29 | 재단법인차세대융합기술연구원 | Control server for estimating traffic density using adas probe data |
| WO2022077767A1 (en) * | 2020-10-16 | 2022-04-21 | 深圳先进技术研究院 | Traffic flow prediction method and apparatus, computer device, and readable storage medium |
| CN112884222A (en) * | 2021-02-10 | 2021-06-01 | 武汉大学 | Time-period-oriented LSTM traffic flow density prediction method |
| CN114333335A (en) * | 2022-03-15 | 2022-04-12 | 成都交大大数据科技有限公司 | Lane-level traffic state estimation method, device and system based on track data |
Non-Patent Citations (2)
| Title |
|---|
| 基于图卷积神经网络的交通流量预测;朱凯利;朱海龙;刘靖宇;石晔琼;王欢;;智能计算机与应用(第06期);第168-170、177页 * |
| 面向季节性时空数据的预测式循环网络及其在城市计算中的应用;张建晋;王韫博;龙明盛;王建民;王海峰;;计算机学报(第02期);第286-302页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115457764A (en) | 2022-12-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN115457764B (en) | Road section traffic density estimation method, device and medium based on vehicle track data | |
| CN107024216B (en) | Intelligent vehicle fusion positioning system and method introducing panoramic map | |
| CN1909022B (en) | Road map manage system | |
| CN112347993B (en) | Expressway vehicle behavior and track prediction method based on vehicle-unmanned aerial vehicle cooperation | |
| EP3690741A2 (en) | Method for automatically evaluating labeling reliability of training images for use in deep learning network to analyze images, and reliability-evaluating device using the same | |
| CN104599502A (en) | Method for traffic flow statistics based on video monitoring | |
| CN110909909A (en) | Short-term traffic flow prediction method based on deep learning and multi-layer spatiotemporal feature map | |
| Albinsson et al. | Design of tyre force excitation for tyre–road friction estimation | |
| CN110188797A (en) | A kind of intelligent automobile method for rapidly testing based on Bayes's optimization | |
| CN108717796A (en) | A kind of geographical ramp self-learning method and system of vehicle | |
| CN109063771A (en) | A kind of method, device and equipment finding vehicle suspicious actions | |
| CN110194041A (en) | The adaptive bodywork height adjusting method of Multi-source Information Fusion | |
| CN110414421A (en) | A kind of Activity recognition method based on sequential frame image | |
| CN116010838A (en) | Vehicle track clustering method integrating density value and K-means algorithm | |
| CN112562311B (en) | Method and device for obtaining working condition weight factor based on GIS big data | |
| CN113362600B (en) | Traffic state estimation method and system | |
| CN116861262B (en) | Perception model training method and device, electronic equipment and storage medium | |
| CN110751850B (en) | Parking space identification method and system based on deep neural network | |
| CN110126846A (en) | Representation method, device, system and the storage medium of Driving Scene | |
| CN116109986A (en) | Vehicle track extraction method based on laser radar and video technology complementation | |
| CN113362488B (en) | OBD monitoring data processing system | |
| CN111582563B (en) | Short-term prediction method, system, device and storage medium for individual travel time | |
| CN110135219A (en) | Data vacuate method, apparatus, storage equipment, map, control system and vehicle | |
| CN114299433A (en) | Abnormal driving behavior detection method, training method, device and storage medium | |
| CN113780157A (en) | Unsupervised reverse detection algorithm combining deep learning and Gaussian mixture model |
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 |