CN112652172B - Road section traffic volume analysis method based on vehicle GPS track - Google Patents

Road section traffic volume analysis method based on vehicle GPS track Download PDF

Info

Publication number
CN112652172B
CN112652172B CN202110068588.4A CN202110068588A CN112652172B CN 112652172 B CN112652172 B CN 112652172B CN 202110068588 A CN202110068588 A CN 202110068588A CN 112652172 B CN112652172 B CN 112652172B
Authority
CN
China
Prior art keywords
road section
road
traffic volume
gps track
vehicle
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
Application number
CN202110068588.4A
Other languages
Chinese (zh)
Other versions
CN112652172A (en
Inventor
张国强
王斯琨
徐炜铃
陈峻
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 CN202110068588.4A priority Critical patent/CN112652172B/en
Publication of CN112652172A publication Critical patent/CN112652172A/en
Application granted granted Critical
Publication of CN112652172B publication Critical patent/CN112652172B/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
    • G—PHYSICS
    • G08—SIGNALLING
    • G08G—TRAFFIC CONTROL SYSTEMS
    • G08G1/00—Traffic control systems for road vehicles
    • G08G1/01—Detecting movement of traffic to be counted or controlled
    • G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
    • G08G1/0137—Measuring and analyzing of parameters relative to traffic conditions for specific applications
    • G—PHYSICS
    • G08—SIGNALLING
    • G08G—TRAFFIC CONTROL SYSTEMS
    • G08G1/00—Traffic control systems for road vehicles
    • G08G1/065—Traffic control systems for road vehicles by counting the vehicles in a section of the road or in a parking area, i.e. comparing incoming count with outgoing count

Landscapes

  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Chemical & Material Sciences (AREA)
  • Analytical Chemistry (AREA)
  • Traffic Control Systems (AREA)

Abstract

The invention discloses a GPS based on a vehicleThe method for analyzing the road section traffic volume of the track comprises the following steps: (1) reading road section information, actually measured traffic volume data and vehicle GPS track data; (2) calculating the shortest path between any two nodes in the road network; (3) inquiring the vehicle type of the GPS track point, and calculating a matched road section of the GPS track point; (4) calculating the traffic volume of various vehicle types with GPS track data on a road section; (5) by usingKDividing road sections by a folding and crossing verification method, constructing a road section traffic volume prediction model, and comparing and selecting the models; (6) and calculating the traffic volume of the road section matched with the GPS track point by using the road section traffic volume prediction model. The invention constructs a traffic prediction model by using the vehicle GPS track data, estimates the traffic of other road sections matched with the GPS track data on the basis of acquiring the traffic of a small number of actually measured road sections, provides abundant and detailed data for traffic planning, management and control and better supports the practical activities of traffic engineering.

Description

Road section traffic volume analysis method based on vehicle GPS track
Technical Field
The invention relates to a road section traffic volume analysis method based on a vehicle GPS track, and belongs to the technical field of traffic engineering.
Background
The analysis of the road traffic volume is the basic work of traffic engineering, and provides basic data for traffic planning, traffic management and traffic control. Whether planning new roads, widening streets, modifying road plane intersections, setting bus lanes, managing roadside parking, optimizing timing of traffic lights, setting pedestrian crossings, or evaluating safety of road traffic, analyzing congestion conditions of traffic flows, and calculating carbon emission of motor vehicles, a large amount of road traffic data is required to be acquired. Therefore, the data acquisition of the road traffic volume is a core technical problem to be solved by traffic engineering.
The traditional original traffic data acquisition method requires a researcher to record the arrival condition of a vehicle on site, needs to consume a lot of time and energy, and is easy to generate various artificial errors, so that the data acquisition quality is not high, and errors are difficult to control. Currently, with the development of electronic and information technologies, various vehicle detection devices such as coil detectors, ultrasonic detectors, geomagnetic detectors, video detectors and the like are widely applied in the practice of traffic engineering, and a convenient and fast way is provided for acquiring traffic data. However, due to the high cost, the road section for installing the automatic vehicle detection device is very limited, and the actual requirements of traffic engineering can not be met.
With the gradual development of the internet of vehicles and Location-Based Services (Location Based Services), the GPS data which records the vehicle time-space information is easy to collect, large in data volume and wide in distribution, and is very beneficial to data mining and research on the vehicle travel characteristics. The vehicle GPS track data provided by the taxi and bus company does not relate to personal privacy, and is convenient to apply. Based on GPS track data of various vehicles, track points are matched to road sections by using a road matching algorithm, and track flow of various vehicle types can be obtained. And establishing a functional relation between the track flow data of the road section and the actually measured traffic volume data, and estimating the traffic volume of other road sections without acquiring the actually measured traffic volume data, so that abundant traffic volume data can be acquired in a large amount at low cost.
In consideration of the limitations of the road traffic data acquisition method and the restriction of high data acquisition cost, the current traffic engineering technical practice needs a road traffic analysis method based on vehicle GPS track data very urgently, so that the limitations of the traffic data acquisition technology are effectively broken through, and accurate and rich traffic information is obtained cheaply and efficiently.
Disclosure of Invention
The technical problem to be solved by the invention is as follows: the method for analyzing the road section traffic volume based on the vehicle GPS track can effectively utilize the vehicle GPS track data to analyze the change rule of the road section traffic volume, construct a road section traffic volume prediction model, calculate the estimated traffic volume of the road section without actually measured traffic volume, and provide data support for practical activities of traffic engineering.
The invention adopts the following technical scheme for solving the technical problems:
a road section traffic volume analysis method based on a vehicle GPS track comprises the following steps:
step 1, obtaining information of each road section on a road network and actually measured traffic volume data of the road section;
step 2, acquiring vehicle GPS track data, and cleaning the vehicle GPS track data;
step 3, calculating and obtaining a shortest path between any two nodes on the road network by using a Floyd shortest path algorithm, wherein the nodes are intersections on the road network;
step 4, calculating the matched road section of each GPS track point for the vehicle GPS track data cleaned in the step 2;
step 5, calculating the traffic volume of various vehicle types with GPS track data on the road section;
step 6, selecting road sections with actually measured traffic volume from the matched road sections of all GPS track points, dividing the road sections by adopting a K-fold cross validation method, constructing a road section traffic volume prediction model by using various methods, and selecting a final road section traffic volume prediction model according to the average error;
and 7, selecting a road section without actually measured traffic volume from the matched road sections of all the GPS track points, and calculating the road section traffic volume by using a final road section traffic volume prediction model.
As a preferred scheme of the present invention, the road section information in step 1 includes a road section number, a road section length, a road section start node, a road section end node, a road section lane number and a single lane traffic capacity; the road section measured traffic data comprises road section numbers, vehicle types and measured traffic.
As a preferable aspect of the present invention, the actually measured traffic volume is manually obtained by means of a traffic survey and/or automatically obtained by a detection device, and the detection device includes a coil detector, an ultrasonic detector, a geomagnetic detector and/or a video detector.
As a preferred embodiment of the present invention, the specific process of step 2 is as follows:
step 21, acquiring vehicle GPS track data, including license plate, serial number, longitude, latitude and positioning time of each GPS track point;
step 22, determining an analysis time range, and acquiring a vehicle GPS track data record in the time range;
step 23, determining the latitude and longitude range of the area where the road network is located: longitude LN of east end of area of road networkEAnd the west longitude LNWRational interval [ LN ] as longitudeW,LNE]The north-most latitude LA of the area of the road networkNAnd southerst latitude LASReasonable interval as latitude [ LAS,LAN]；
Step 24, extracting longitude lng of ith GPS track point of vehicleiAnd latitude lati；
Step 25, if ngiAnd latiIf at least one of the GPS track points is a non-number, deleting the data record of the GPS track point;
step 26, if ngi>LNEOr lngi<LNWDeleting the data record of the GPS track point;
step 27, if lati>LANOr lati<LASAnd deleting the data record of the GPS track point.
As a preferred embodiment of the present invention, the specific process of step 3 is as follows:
step 31, selecting any node t on a road network;
step 32, selecting any two nodes r and s on the road network;
in step 33, for the distance Dist (r, s) from the node r to the node s, the distance Dist (r, t) from the node r to the node t, and the distance Dist (t, s) from the node t to the node s, if the following conditions are satisfied: dist (r, s) > Dist (r, t) + Dist (t, s), let: dist (r, s) — Dist (r, t) + Dist (t, s), and Path (r, s) — t, i.e., the Path (r, s) from node r to node s includes node t.
As a preferred embodiment of the present invention, the specific process of step 4 is as follows:
step 41, for the ith GPS track point p of the vehicleiExtracting the number numb of the cariLongitude and latitude coordinates (lng)i,lati)；
Step 42, inquiring number numb of license plateiThe corresponding vehicle type;
step 43, let the GPS error radius be R, find out the tracing point piAll road sections covered by a circle with the circle center as R as the radius are obtained to obtain a candidate road section set omegai；
Step 44, for any candidate road segment segj∈ΩiCalculating the trace point piShortest distance d to the candidate linkij: if the locus point piIs located above the candidate road section, the shortest distance is the track point piA vertical segment length to the candidate segment; if the locus point piIs located outside the candidate road section, the shortest distance is the track point piThe minimum value of the distances to the starting node and the ending node of the candidate road section;
step 45, calculating the trace point piMinimum value of shortest distances to all candidate links
Figure BDA0002905139460000041
The minimum value diThe corresponding candidate road section is the track point piThe matching road section of (1).
As a preferred embodiment of the present invention, the specific process of step 5 is as follows:
step 51, for the license plate number nb with the GPS track data, the corresponding vehicle type is set to be tp, and a GPS track point set is established as traj according to the GPS track datanb＝{p1,p2,…,pzThe elements in the set are arranged according to time sequence, p1,p2,…,pzAll are trace points;
step 52, in the GPS track point set trajnbFrom the n-th1Start of an element, n11,2, …, z, finding the matching road segment corresponding to it
Figure BDA0002905139460000042
All being other points of the same road section
Figure BDA0002905139460000043
Up to the matching section corresponding thereto
Figure BDA0002905139460000044
Not of the same road section
Figure BDA0002905139460000045
Occurrence of n2Not less than 1, i.e. matching road sections
Figure BDA0002905139460000046
And matching road sections
Figure BDA0002905139460000047
All of which are identical but matching
Figure BDA0002905139460000048
And matching road sections
Figure BDA0002905139460000049
Not the same road segment; if n is2>1, go to step 53, if n2If 1, go to step 54;
step 53, matching road sections
Figure BDA00029051394600000410
Let vol be the same road section, let it be road section uu,tp＝volu,tp+1，n1＝n1+n2And returning to step 52, volu,tpRepresenting the traffic volume of a vehicle of type tp on a road section u;
step 54, query matching road section
Figure BDA00029051394600000411
To matching road section
Figure BDA00029051394600000412
The shortest path Route of (a) is,
Figure BDA00029051394600000413
in the form of a tail section of the path,
Figure BDA00029051394600000414
for the head section of the path, let volu,tp＝volu,tp+1，
Figure BDA00029051394600000415
n1＝n1+1 and returns to step 52.
As a preferred embodiment of the present invention, the specific process of step 6 is as follows:
step 61, selecting road sections with actually measured traffic volume from the matched road sections of all GPS track points, and constructing a road section set ANA for data analysis;
step 62, for any road section segma ∈ ANA, the actually measured traffic volume y of the road section is calculatedsegmaTraffic volume vol of a vehicle type with GPS trajectory data on a road section as an output variablesegmaNumber of lanes lan on road sectionsegmaAnd single lane capacity capsegmaAs input variables;
step 63, the road section set ANA has nsDividing a road section set ANA by adopting a K-fold cross validation method for each road section: disordering the elements in the road section set ANA, and taking out ntestIndividual road section as test set ANAtestN will remains-ntestDividing each road section into K parts in equal proportion, taking 1 part as a verification set, taking the rest K-1 parts as a training set, repeating for K times, and adopting different verification sets each time;
step 64, building a road section traffic volume prediction model by respectively applying H methods, and building a functional relation y between an input variable and an output variablesegma＝fh(volsegma,lansegma,capsegma)，h＝1,2,…,H；
Step 65, for the prediction model fhThe average error of K verifications was calculated:
Figure BDA0002905139460000051
Figure BDA0002905139460000052
wherein, TESTkDenotes the kth verification set, ysegmaRepresents the measured traffic volume of the section segma,
Figure BDA0002905139460000053
representing a prediction model fhPredicted traffic volume for the segment segma;
step 66, calculate
Figure BDA0002905139460000054
F is thenminThe final road traffic prediction model is obtained.
As a preferred embodiment of the present invention, the specific process of step 7 is as follows:
step 71, selecting road sections without actually measured traffic volume from the matched road sections of all GPS track points, and constructing a road section set FST for traffic volume prediction;
step 72, for any road section segmf epsilon FST, obtaining the traffic volume vol of the vehicle type with the GPS track data on the road sectionsegmfNumber of lanes lan on road sectionsegmfAnd single lane capacity capsegmf；
Step 73, applying the final road section traffic volume prediction model fminCalculating the traffic volume of the road section segmf: y issegmf＝fmin(volsegmf,lansegmf,capsegmf)。
Compared with the prior art, the invention adopting the technical scheme has the following technical effects:
1. according to the invention, the road network information and the vehicle GPS track data are organically fused by acquiring and processing the road network information and the vehicle GPS track data and matching the GPS track point with the road section, so that the traffic volume of various vehicle types with the GPS track data on the road section is acquired on the basis, the vehicle GPS track data which is huge in quantity, wide in distribution and disordered is efficiently integrated, and the use value of the data is greatly improved.
2. The invention analyzes the function relation between the actually measured traffic volume of the road section and the traffic volume of the vehicle type with the GPS track data on the road section and other related factors by using various methods, constructs a road section traffic volume prediction model based on the vehicle GPS track data, and provides a new way and a method for acquiring the road section traffic volume.
3. According to the change rule of the road traffic volume, the road traffic volume analysis based on the vehicle GPS track is realized, rich traffic information contained in mass GPS track data can be deeply and fully mined, rich, detailed and reliable traffic data are provided on the basis of not increasing equipment expenses, and the method has important social and economic significance.
Drawings
FIG. 1 is a flow chart of a road traffic volume analysis method based on a vehicle GPS track according to the present invention.
FIG. 2 is a diagram of a taxi GPS track point distribution on a road network of a certain county area.
FIG. 3 is a diagram of a GPS track point distribution of a bus on a road network in a certain county area.
Detailed Description
Reference will now be made in detail to embodiments of the present invention, examples of which are illustrated in the accompanying drawings. The embodiments described below with reference to the accompanying drawings are illustrative only for the purpose of explaining the present invention, and are not to be construed as limiting the present invention.
The method organically fuses abundant vehicle GPS track data, road network information and actual measurement traffic data of the road section with limited quantity, constructs a road section traffic prediction model, and calculates the large quantity of road section traffic without the actual measurement data on the basis.
The present invention will be described in further detail with reference to the accompanying drawings. Fig. 1 shows a flow chart of the present invention, which specifically includes the following steps:
1. and opening a road network file consisting of road sections, and reading information of each road section and actually measured traffic volume data of the road sections. The road section information comprises road section numbers, road section lengths, road section starting nodes, road section ending nodes, road section lane numbers and single lane traffic capacity; the road section actual measurement traffic data comprises road section numbers, vehicle types and traffic volumes, and the actual measurement traffic data can be manually acquired in a traffic investigation mode and can also be automatically acquired through various detection devices such as a coil detector, an ultrasonic detector, a geomagnetic detector and a video detector.
2. And reading vehicle GPS track data, including information such as the serial number, license plate, longitude, latitude, positioning time and the like of each track point. The method for cleaning the vehicle GPS track data specifically comprises the following steps:
(1) determining an analyzed time range, and reading vehicle GPS track data in the time range;
(2) determining the latitude and longitude range of the area where the road network is located: longitude LN of the east and west ends of the regionEAnd LNWRational interval [ LN ] as longitudeW,LNE]Latitude LA at the north and south extremes of the area to be studiedNAnd LASReasonable interval as latitude [ LAS,LAN]；
(3) For the ith vehicle GPS track point, extracting the longitude coordinate lng thereofiAnd latitude coordinate lati；
(4) If ngiIs not a number, or latiIs not a number; deleting the record;
(5) if inequality lngi>LNEIs true, or is inequality lngi<LNWIf true; deleting the record;
(6) if inequality lati>LANOr a true or an inequality lati<LASIf true; the record is deleted.
Fig. 2 and fig. 3 are GPS track point distribution diagrams of taxies and buses on a road network of a certain county area respectively.
3. The shortest circuit between any two points in the circuit network is calculated by using a Floyd shortest circuit algorithm, and the calculation result is stored, specifically:
(31) selecting any node t in a road network;
(32) selecting any two nodes r and s in a road network;
(33) for the distance Dist (r, s) from node r to s, the distance Dist (r, t) from node r to node t, and the distance Dist (t, s) from node t to node s, if the following conditions are satisfied: dist (r, s) > Dist (r, t) + Dist (t, s), let: dist (r, s) ═ Dist (r, t) + Dist (t, s), and Path (r, s) ═ t.
4. Inquiring the vehicle type of the GPS track point, and calculating the matching road section of the GPS track point, which specifically comprises the following steps:
(41) for the ith vehicle GPS track point, the number plate numb is extractediLongitude and latitude coordinate pi＝(lngi,lati)；
(42) Enquiry of number nb of vehicleiThe corresponding vehicle type;
(43) let the error radius of GPS be R, find out the tracing point piObtaining a candidate road section set omega for the road section covered by the circle with the circle center R as the radiusi；
(44) Seg for any one candidate linkj∈ΩiCalculating the trace point piShortest distance d to the road sectionij: if the locus point piThe foot of the user is positioned above the road section, and the shortest distance is the track point piLength of vertical segment to road segment; if the locus point piIs located outside the road section, the shortest distance is the track point piA minimum of distances to two end points of the road section;
(45) calculating and calculating the trace point piMinimum value of shortest distances to all candidate links
Figure BDA0002905139460000071
The minimum value diThe corresponding candidate road section is the track point piThe matching road section of (1).
5. Calculating the traffic volume of various vehicle types with GPS track data on the road section, specifically:
(51) setting the corresponding vehicle type as tp for the license plate number nb with the GPS track, and constructing a GPS track point set trajnb＝{p1,p2,…,pzArranging the elements in the set according to the time sequence;
(52) at GPS track point set trajnbFrom n to1(n11,2, …, z) elements, and continuously selecting a plurality of track points
Figure BDA0002905139460000081
Searching the corresponding matching road section
Figure BDA0002905139460000082
The conditions are satisfied: matching road section
Figure BDA0002905139460000083
And matching road sections
Figure BDA0002905139460000084
All of which are the same road segment but different from the matching road segment
Figure BDA0002905139460000085
(53) If n is2>1, then matching the road section
Figure BDA0002905139460000086
For the same road section, let it be road section u, let volu,tp＝volu,tp+1，n1＝n1+n2And back (52); volu,tpRepresenting the traffic volume of a vehicle of type tp on a road section u;
(54) if n is2Query by matching road segment 1
Figure BDA0002905139460000087
To matching road section
Figure BDA0002905139460000088
Of (2) a
Figure BDA0002905139460000089
In the form of a tail section of the path,
Figure BDA00029051394600000810
is a path head section; let volu,tp＝volu,tp+1,
Figure BDA00029051394600000811
n1＝n1+1, and back (52); volu,tpRepresenting the traffic volume of a vehicle of type tp on a road section u.
6. Dividing road sections by adopting a K-fold cross validation method, constructing a road section traffic volume prediction model by using various methods, and comparing and selecting the models according to the average error, wherein the method specifically comprises the following steps:
(61) selecting a road section with actually measured traffic volume from road sections matched with the GPS track points, and constructing a road section set ANA for data analysis;
(62) for any road section segma ∈ ANA, the actually measured traffic volume of the road section is used as an output variable ysegmaVolume of traffic on road section of vehicle type with GPS trajectory datasegmaNumber of lanes lan on road sectionsegmaTraffic capacity cap of single lanesegmaEtc. to form input variables;
(63) the road section set ANA has nsDividing a road section set ANA by adopting a K-fold cross validation method for each road section: disordering the elements in the road section set ANA, and taking out ntestIndividual road section as test set ANAtestNot used for training the model, but only used for final model error evaluation; n will remains-ntestDividing each road section into K parts in equal proportion, taking 1 part as a verification set, taking the rest K-1 parts as a training set, repeating for K times, and adopting different verification sets each time to ensure that all K parts of samples are tested;
(64) and (3) constructing a prediction model by respectively applying H methods, such as: linear regression, support vector regression, BP neural network, etc., establishing the functional relationship y between input variable and output variablesegma＝fh(volsegma,lansegma,capsegma)(h＝1,2,…,H)；
(65) For the prediction model fh(H ═ 1,2, …, H), the average error for its K verifications was calculated:
Figure BDA0002905139460000091
wherein, TESTkDenotes the kth verification set, segma denotes a road segment, ysegmaThe measured amount of traffic is represented as,
Figure BDA0002905139460000093
representation model fhPredicting the traffic volume;
(66) computing
Figure BDA0002905139460000092
F is thenminThe model that is finally selected.
7. Calculating the road traffic volume with GPS track data by using a road traffic volume prediction model, which specifically comprises the following steps:
(71) selecting a road section without actually measured traffic volume from the road sections matched with the GPS track points, and constructing a road section set FST for traffic volume prediction;
(72) for any road segment segmf epsilon FST, acquiring the traffic volume vol of the vehicle type with the GPS track data on the road segmentsegmfNumber of lanes lan on road sectionsegmfTraffic capacity cap of single lanesegmfAnd the like;
(73) using a prediction model fminCalculating the traffic volume of the road section segmf: y isssegmf＝fmin(volsegmf,lansegmf,capsegmf)。
The above embodiments are only for illustrating the technical idea of the present invention, and the protection scope of the present invention is not limited thereby, and any modifications made on the basis of the technical scheme according to the technical idea of the present invention fall within the protection scope of the present invention.

Claims (7)

1. A road section traffic volume analysis method based on a vehicle GPS track is characterized by comprising the following steps:
step 1, obtaining information of each road section on a road network and actually measured traffic volume data of the road section;
step 2, acquiring vehicle GPS track data, and cleaning the vehicle GPS track data;
step 3, calculating and obtaining a shortest path between any two nodes on the road network by using a Floyd shortest path algorithm, wherein the nodes are intersections on the road network;
step 4, calculating the matched road section of each GPS track point for the vehicle GPS track data cleaned in the step 2;
the specific process of the step 4 is as follows:
step 41, for the ith GPS track point p of the vehicleiExtracting the license plate number nb and longitude and latitude coordinates (lng)i，lati)；
Step 42, inquiring the vehicle type corresponding to the license plate number nb;
step 43, let the GPS error radius be R, find out the tracing point piAll road sections covered by a circle with the circle center as R as the radius are obtained to obtain a candidate road section set omegai；
Step 44, for any candidate road segment segj∈ΩiCalculating the trace point piShortest distance d to the candidate linkij: if the locus point piIs located above the candidate road section, the shortest distance is the track point piA vertical segment length to the candidate segment; if the locus point piIs located outside the candidate road section, the shortest distance is the track point piThe minimum value of the distances to the starting node and the ending node of the candidate road section;
step 45, calculating the trace point piMinimum value of shortest distances to all candidate links
Figure FDA0003295228360000011
The minimum value diThe corresponding candidate road section is the track point piThe matching road section of (1);
step 5, calculating the traffic volume of various vehicle types with GPS track data on the road section;
the specific process of the step 5 is as follows:
step 51, for the license plate number nb with the GPS track data, the corresponding vehicle type is set to be tp, and a GPS track point set is established as traj according to the GPS track datanb＝{p1，p2，…，pzThe elements in the set are arranged according to time sequence, p1，p2，…，pzAll are trace points;
step 52, in the GPS track point set trajnbFrom the n-th1Start of an element, n11,2, …, z, finding the matching road segment corresponding to it
Figure FDA0003295228360000012
All being other points of the same road section
Figure FDA0003295228360000013
Up to the matching section corresponding thereto
Figure FDA0003295228360000021
Not of the same road section
Figure FDA0003295228360000022
Occurrence of n2Not less than 1, i.e. matching road sections
Figure FDA0003295228360000023
And matching road sections
Figure FDA0003295228360000024
All of which are identical but matching
Figure FDA0003295228360000025
And matching road sections
Figure FDA0003295228360000026
Not the same road segment; if n is2>1, go to step 53, if n2If 1, go to step 54;
step 53, matching road sections
Figure FDA0003295228360000027
Is the same waySection, if it is road section u, let volu,tp＝volu,tp+1，n1＝n1+n2-1 and return to step 52, volu,tpRepresenting the traffic volume of a vehicle of type tp on a road section u;
step 54, query matching road section
Figure FDA0003295228360000028
To matching road section
Figure FDA0003295228360000029
The shortest path Route of (a) is,
Figure FDA00032952283600000210
in the form of a tail section of the path,
Figure FDA00032952283600000211
for the head section of the path, let volu，tp＝volu，tp+1，
Figure FDA00032952283600000212
n1＝n1+1 and return to step 52;
step 6, selecting road sections with actually measured traffic volume from the matched road sections of all GPS track points, dividing the road sections by adopting a K-fold cross validation method, constructing a road section traffic volume prediction model by using various methods, and selecting a final road section traffic volume prediction model according to the average error;
and 7, selecting a road section without actually measured traffic volume from the matched road sections of all the GPS track points, and calculating the road section traffic volume by using a final road section traffic volume prediction model.
2. The method for analyzing road section traffic volume based on vehicle GPS track according to claim 1, characterized in that, the road section information of step 1 comprises road section number, road section length, road section starting node, road section ending node, road section lane number and single lane traffic capacity; the road section measured traffic data comprises road section numbers, vehicle types and measured traffic.
3. The method for analyzing road traffic volume based on vehicle GPS track according to claim 2, characterized in that the measured traffic volume is obtained manually by means of traffic investigation and/or automatically by means of detection devices, including coil detectors, ultrasonic detectors, geomagnetic detectors and/or video detectors.
4. The road section traffic volume analysis method based on the vehicle GPS track according to claim 1, characterized in that the specific process of the step 2 is as follows:
step 21, acquiring vehicle GPS track data, including license plate, serial number, longitude, latitude and positioning time of each GPS track point;
step 22, determining an analysis time range, and acquiring a vehicle GPS track data record in the time range;
step 23, determining the latitude and longitude range of the area where the road network is located: longitude LN of east end of area of road networkEAnd the west longitude LNWRational interval [ LN ] as longitudeW,LNE]The north-most latitude LA of the area of the road networkNAnd southerst latitude LASReasonable interval as latitude [ LAS,LAN]；
Step 24, extracting longitude lng of ith GPS track point of vehicleiAnd latitude lati；
Step 25, if ngiAnd latiIf at least one of the GPS track points is a non-number, deleting the data record of the GPS track point;
step 26, if ngi>LNEOr lngi<LNWDeleting the data record of the GPS track point;
step 27, if lati>LANOr lati<LAsAnd deleting the data record of the GPS track point.
5. The road section traffic volume analysis method based on the vehicle GPS track according to claim 1, characterized in that the specific process of the step 3 is as follows:
step 31, selecting any node t on a road network;
step 32, selecting any two nodes r and s on the road network;
in step 33, for the distance Dist (r, s) from the node r to the node s, the distance Dist (r, t) from the node r to the node t, and the distance Dist (t, s) from the node t to the node s, if the following conditions are satisfied: dist (r, s) > Dist (r, t) + Dist (t, s), let: dist (r, s) — Dist (r, t) + Dist (t, s), and Path (r, s) — t, i.e., the Path (r, s) from node r to node s includes node t.
6. The method for analyzing the road traffic volume based on the vehicle GPS track according to claim 1, characterized in that the specific process of the step 6 is as follows:
step 61, selecting road sections with actually measured traffic volume from the matched road sections of all GPS track points, and constructing a road section set ANA for data analysis;
step 62, for any road section segma ∈ ANA, the actually measured traffic volume y of the road section is calculatedsegmaTraffic volume vol of a vehicle type with GPS trajectory data on a road section as an output variablesegmaNumber of lanes lan on road sectionsegmaAnd single lane capacity capsegmaAs input variables;
step 63, the road section set ANA has nsDividing a road section set ANA by adopting a K-fold cross validation method for each road section: disordering the elements in the road section set ANA, and taking out ntestIndividual road section as test set ANAtestN will remains-ntestDividing each road section into K parts in equal proportion, taking 1 part as a verification set, taking the rest K-1 parts as a training set, repeating for K times, and adopting different verification sets each time;
step 64, building a road section traffic volume prediction model by respectively applying H methods, and building a functional relation y between an input variable and an output variablesegma＝fh(volsegma,lansegma，capsegma)，h＝1,2,…,H；
Step 65, for the prediction model fhThe average error of K verifications was calculated:
Figure FDA0003295228360000041
Figure FDA0003295228360000042
wherein, TESTkDenotes the kth verification set, ysegmaRepresents the measured traffic volume of the section segma,
Figure FDA0003295228360000043
representing a prediction model fhPredicted traffic volume for the segment segma;
step 66, calculate
Figure FDA0003295228360000044
F is thenminThe final road traffic prediction model is obtained.
7. The method for analyzing the road traffic volume based on the vehicle GPS track according to claim 1, characterized in that the specific process of the step 7 is as follows:
step 71, selecting road sections without actually measured traffic volume from the matched road sections of all GPS track points, and constructing a road section set FST for traffic volume prediction;
step 72, for any road section segmf epsilon FST, obtaining the traffic volume vol of the vehicle type with the GPS track data on the road sectionsegmfNumber of lanes lan on road sectionsegmfAnd single lane capacity capsegmf；
Step 73, applying the final road section traffic volume prediction model fminCalculating the traffic volume of the road section segmf: y issegmf＝fmin(volsegmf，lansegmf，capsegmf)。
CN202110068588.4A 2021-01-19 2021-01-19 Road section traffic volume analysis method based on vehicle GPS track Active CN112652172B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202110068588.4A CN112652172B (en) 2021-01-19 2021-01-19 Road section traffic volume analysis method based on vehicle GPS track

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202110068588.4A CN112652172B (en) 2021-01-19 2021-01-19 Road section traffic volume analysis method based on vehicle GPS track

Publications (2)

Publication Number Publication Date
CN112652172A CN112652172A (en) 2021-04-13
CN112652172B true CN112652172B (en) 2022-01-25

Family

ID=75370545

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202110068588.4A Active CN112652172B (en) 2021-01-19 2021-01-19 Road section traffic volume analysis method based on vehicle GPS track

Country Status (1)

Country Link
CN (1) CN112652172B (en)

Families Citing this family (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113990075B (en) * 2021-12-30 2022-03-18 广州市交通规划研究院 Traffic survey data and track data fused flow distribution method and system
CN114566053B (en) * 2022-03-03 2023-05-16 中交第一公路勘察设计研究院有限公司 Expressway traffic volume prediction optimization algorithm
CN116303835B (en) * 2023-05-24 2023-07-21 北京建筑大学 Mapping data management method for road planning
CN118246835B (en) * 2024-05-21 2024-09-20 天津市政工程设计研究总院有限公司 Setting boundary demarcating method for intelligent network connection special channel

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106384504A (en) * 2016-09-06 2017-02-08 江苏智通交通科技有限公司 Urban road network jam feature description analysis method based on data visualization
CN108197724A (en) * 2017-12-02 2018-06-22 山东大学 Efficiency weight calculation and public transport scheme performance estimating method in public transport complex network
CN109300310A (en) * 2018-11-26 2019-02-01 平安科技（深圳）有限公司 A kind of vehicle flowrate prediction technique and device
CN111696355A (en) * 2020-06-29 2020-09-22 中南大学 Dynamic graph convolution traffic speed prediction method
CN111862594A (en) * 2020-06-04 2020-10-30 中国兵器科学研究院 Method, device and storage medium for identifying weak unit in road traffic network

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106384504A (en) * 2016-09-06 2017-02-08 江苏智通交通科技有限公司 Urban road network jam feature description analysis method based on data visualization
CN108197724A (en) * 2017-12-02 2018-06-22 山东大学 Efficiency weight calculation and public transport scheme performance estimating method in public transport complex network
CN109300310A (en) * 2018-11-26 2019-02-01 平安科技（深圳）有限公司 A kind of vehicle flowrate prediction technique and device
CN111862594A (en) * 2020-06-04 2020-10-30 中国兵器科学研究院 Method, device and storage medium for identifying weak unit in road traffic network
CN111696355A (en) * 2020-06-29 2020-09-22 中南大学 Dynamic graph convolution traffic speed prediction method

Also Published As

Publication number Publication date
CN112652172A (en) 2021-04-13

Similar Documents

Publication Publication Date Title
CN112652172B (en) Road section traffic volume analysis method based on vehicle GPS track
CN110111575B (en) Urban traffic flow network analysis method based on complex network theory
US7516041B2 (en) System and method for identifying road features
Yuan et al. Driving with knowledge from the physical world
Gentili et al. Review of optimal sensor location models for travel time estimation
Kuchipudi et al. Development of a hybrid model for dynamic travel-time prediction
Balakrishna et al. Time-dependent origin-destination estimation without assignment matrices
Yao et al. An optimization model for arterial coordination control based on sampled vehicle trajectories: The STREAM model
CN108921383B (en) Method and system for acquiring urban road traffic redundant carbon emission index
Carisi et al. Enhancing in vehicle digital maps via GPS crowdsourcing
Rupi et al. Evaluating cyclist patterns using GPS data from smartphones
CN106709196B (en) Distribution method of motor vehicle exhaust remote measuring equipment based on graph theory
CN113724489A (en) Traffic jam tracing method based on multi-source data
CN116597649B (en) Road network traffic flow deduction method based on expressway charging system
Liu et al. Themis: A participatory navigation system for balanced traffic routing
Zhao et al. Development and validation of improved impedance functions for roads with mixed traffic using taxi GPS trajectory data and simulation
CN114973657A (en) Urban traffic noise pollution analysis and evaluation method based on trajectory data
Ghandeharioun et al. Link travel time estimation for arterial networks based on sparse GPS data and considering progressive correlations
Nadeem et al. Performance analysis of a real-time adaptive prediction algorithm for traffic congestion
CN111710160A (en) Travel time prediction method based on floating car data
Gore et al. Exploring credentials of Wi‐Fi sensors as a complementary transport data: an Indian experience
CN110675631A (en) Traffic flow traceability analysis method and system
CN112183871A (en) Urban traffic guidance system based on air index
El Esawey et al. Calibration and validation of micro-simulation models of medium-size networks.
Efentakis et al. Crowdsourcing turning restrictions for OpenStreetMap.

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