CN116383678A - Method for identifying abnormal speed change behavior frequent road sections of operating passenger car - Google Patents

Method for identifying abnormal speed change behavior frequent road sections of operating passenger car Download PDF

Info

Publication number
CN116383678A
CN116383678A CN202211578800.2A CN202211578800A CN116383678A CN 116383678 A CN116383678 A CN 116383678A CN 202211578800 A CN202211578800 A CN 202211578800A CN 116383678 A CN116383678 A CN 116383678A
Authority
CN
China
Prior art keywords
speed change
acceleration
abnormal speed
abnormal
road
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN202211578800.2A
Other languages
Chinese (zh)
Other versions
CN116383678B (en
Inventor
别一鸣
张国庆
刘亚君
肖乔云
王天贺
李振宁
朱奥泽
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.)
Jilin University
Original Assignee
Jilin 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 Jilin University filed Critical Jilin University
Priority to CN202211578800.2A priority Critical patent/CN116383678B/en
Publication of CN116383678A publication Critical patent/CN116383678A/en
Application granted granted Critical
Publication of CN116383678B publication Critical patent/CN116383678B/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
    • G08G1/052—Detecting movement of traffic to be counted or controlled with provision for determining speed or overspeed
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
    • G06F16/29—Geographical information databases
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/22—Matching criteria, e.g. proximity measures
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/23—Clustering techniques
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/29—Graphical models, e.g. Bayesian networks
    • G06F18/295—Markov models or related models, e.g. semi-Markov models; Markov random fields; Networks embedding Markov models
    • G06Q50/40—
    • 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 method for identifying frequent road sections of abnormal speed change behaviors of a passenger car, and relates to a method for identifying frequent road sections of abnormal speed change behaviors of a passenger car. The invention aims to solve the problems that the existing traffic safety hidden danger road section investigation method belongs to statistical analysis after accident occurrence, hidden danger road sections are not recognized and early-warned before accident occurrence, and the economic cost is high and the precision is low. The method for identifying the frequent road sections of the abnormal speed change behavior of the operating passenger car comprises the following specific processes: step 1, basic data acquisition; step 2, correcting the coordinates; step 3, calculating basic data; step 4, identifying abnormal speed change behaviors; and 5, identifying the frequently abnormal speed change road sections. The invention belongs to the technical field of public transportation operation management.

Description

Method for identifying abnormal speed change behavior frequent road sections of operating passenger car
Technical Field
The invention relates to a method for identifying a common road section of abnormal speed change behavior of a commercial passenger car. The invention belongs to the technical field of public transportation operation management.
Background
As an important public transportation mode, the commercial passenger car has the characteristics of large passenger capacity, high transportation efficiency, comfortable riding environment and the like, and is relatively suitable for middle-short distance travel. Abnormal speed change behaviors such as sudden brake stepping or sudden accelerator stepping of a passenger car driver in the driving process can cause discomfort or injury of passengers, and accidents such as rear-end collision, side turning and the like can also occur, so that more serious personal casualties and property loss are caused. The abnormal speed change behavior on the actual road often has certain space distribution regularity, the road section where the abnormal speed change behavior is frequently generated has higher potential safety hazard, and the identification of the road section where the abnormal speed change behavior is frequently generated of the operating passenger car can timely early warn a driver on one hand, so that traffic accidents are avoided; on the other hand, the common road section information can be provided for a road management department to reform the hidden danger road section so as to improve the driving safety of the road.
The current abnormal behavior judgment of the vehicle mostly adopts a video monitoring or multi-sensor data fusion method, the required data has higher scale and quality requirements, and the economic cost of acquisition equipment and storage equipment is higher. The method has less research on the identification of the road sections with frequent abnormal behaviors of the vehicle, and the related research is to judge the accident risk road sections on the basis of the existing traffic accident data, belongs to statistical analysis after the accident, and does not identify and early warn the hidden trouble road sections before the accident occurs.
The commercial buses have the characteristics of fixed lines, dense shifts and high operation similarity among shifts, and the current commercial buses are all provided with global positioning systems (Global Positioning System, GPS), so that public buses can obtain data such as positions, speeds, accelerations and the like when a large number of commercial buses run on the fixed lines. The abnormal movement behavior of the vehicle can be reflected in the GPS data, if the identification of the road section frequently suffering from the abnormal speed change behavior can be carried out based on the GPS data, other equipment is not required to be additionally purchased for data acquisition, so that the economic cost can be saved, and the accumulated large amount of data can be conveniently utilized for checking the road hidden danger road section.
Disclosure of Invention
The invention aims to solve the problems that the existing traffic safety hidden danger road section investigation method belongs to statistical analysis after accident occurrence, hidden danger road sections are not recognized and early-warned before accident occurrence, the economic cost is high and the accuracy is low, and provides a method for recognizing frequent road sections of abnormal speed change behaviors of a commercial passenger car so as to improve the accuracy and reliability of a frequent road section recognition technology.
The method for identifying the frequent road sections of the abnormal speed change behavior of the operating passenger car comprises the following specific processes:
step 1, basic data acquisition;
step 2, carrying out coordinate deviation correction based on the basic data;
step 3, correcting and calculating the speed and the acceleration based on the basic data and the coordinates;
step 4, identifying abnormal speed change behaviors based on the speed and the acceleration;
and 5, carrying out frequent abnormal speed change road section identification based on the abnormal speed change behavior.
The beneficial effects of the invention are as follows:
according to the method for identifying the frequently-occurring road sections of the abnormal speed change behavior of the commercial passenger car, disclosed by the invention, the coordinate deviation correction is completed by means of map matching by virtue of the vehicle-mounted GPS data which is easy to acquire, so that the negative influence caused by the hardware error of the acquisition equipment is reduced, and the accuracy of the operation basic parameters of the commercial passenger car is improved; determining acceleration threshold values corresponding to the speeds by fitting an acceleration threshold curve so as to judge abnormal speed change behaviors of the operating passenger car; and further, the recognition of frequent road sections of frequent abnormal speed change behaviors is realized based on a density-based noise space clustering method (DBSCAN). The technology can provide road safety pre-warning for operating passenger car drivers, and avoid traffic accidents in road sections frequently caused by abnormal speed change behaviors; meanwhile, the potential safety hazard road section position is provided for the road management department, and assistance is provided for road section reconstruction.
Drawings
FIG. 1 is a flow chart of the present invention.
Detailed Description
The first embodiment is as follows: the method for identifying the frequently-occurring road sections of the abnormal speed change behavior of the operating passenger car comprises the following specific processes:
step 1, basic data acquisition;
step 2, carrying out coordinate deviation correction based on the basic data;
step 3, correcting and calculating the speed and the acceleration based on the basic data and the coordinates;
step 4, identifying abnormal speed change behaviors based on the speed and the acceleration;
and 5, carrying out frequent abnormal speed change road section identification based on the abnormal speed change behavior.
The second embodiment is as follows: the first difference between the present embodiment and the specific embodiment is that in the step 1, basic data is collected; the specific process is as follows:
step 1.1, extracting GPS data of all operating shifts in the past continuous N days of a commercial bus line, wherein the GPS data comprises acquisition time, sampling interval, longitude and latitude, vehicle ID and shift direction information, and the suggested N value is 180;
step 1.2, the space distribution of the frequently-occurring road sections of the abnormal speed change behavior in the up-down direction of the operating passenger car has obvious difference, and the up-down direction operating road of the operating passenger car is selected as a research object, so that the method is also applicable to the down-direction operating road;
step 1.3, defining an operation shift set of the uplink direction of the line as i= { I: i=1, 2, 3., I }; shift i has K GPS data acquisition points in total, and the K data acquisition point coordinates of shift i are expressed as
Figure BDA0003980834580000021
Wherein lng and lat respectively represent longitude and latitude, and the unit is degree; the GPS sampling interval is expressed as T, and the unit is s; k=1, 2,3,;
and 1.4, extracting a GIS road network of a running line passing region of the operating passenger car, wherein the GIS road network comprises a road name, a road number and road grade information.
Other steps and parameters are the same as in the first embodiment.
And a third specific embodiment: the difference between the embodiment and the specific embodiment is that, in the step 2, coordinate rectification is performed based on the basic data (because the acquisition of the GPS coordinates is biased, it cannot be guaranteed that the coordinate points fall on the road network accurately, for the sake of more accuracy, the coordinates after map matching are considered to be more accurate, and here, the conversion of the map matching into planar coordinates is more convenient for subsequent calculation); the specific process is as follows:
2.1, adopting a map matching algorithm based on a hidden Markov model to finish map matching of the GPS data acquisition points and the GIS road network, wherein the matched road network is the road network where the GPS data acquisition points are actually located;
step 2.2. The coordinates of the GPS data acquisition points after map matching are the actual positions of the GPS data acquisition points on the road network, and are expressed as
Figure BDA0003980834580000031
Step 2.3. Matching the coordinates of the map
Figure BDA0003980834580000032
Unified conversion into planar coordinates->
Figure BDA0003980834580000033
(for example, the coordinates after map matching can be uniformly converted into plane coordinates by adopting a projection mode, the projection mode can be a UTM 3-degree banded projection mode), x and y are respectively the projection distances from the GPS data acquisition point to the central meridian and the equator in a coordinate system taking the intersection point of the equator and the central meridian as the origin of coordinates, and the unit is m.
Other steps and parameters are the same as in the first or second embodiment.
The specific embodiment IV is as follows: the difference between the present embodiment and one to three embodiments is that the velocity and the acceleration are calculated based on the basic data and the coordinate deviation correction in the step 3; the specific process is as follows:
step 3.1, the speed after correction of the k data acquisition point coordinates of shift i is expressed as
Figure BDA0003980834580000034
The unit is m/s; calculation speed->
Figure BDA0003980834580000035
The formula is shown as follows;
Figure BDA0003980834580000036
wherein:
Figure BDA0003980834580000037
the x-axis coordinates of plane coordinates after correction of the k+1 data acquisition point coordinates and the k data acquisition point coordinates of shift i are respectively given in m;
Figure BDA0003980834580000038
the y-axis coordinates of plane coordinates after correction of the k+1 data acquisition point coordinates and the k data acquisition point coordinates of shift i are respectively given in m;
step 3.2, the acceleration after correction of the k data acquisition point coordinates of shift i is expressed as
Figure BDA0003980834580000039
In m/s 2 The method comprises the steps of carrying out a first treatment on the surface of the Calculate acceleration->
Figure BDA0003980834580000041
The formula is shown as follows;
Figure BDA0003980834580000042
wherein:
Figure BDA0003980834580000043
the speed after correction is obtained for the k-1 data acquisition point coordinates of shift i.
Other steps and parameters are the same as in one to three embodiments.
Fifth embodiment: the first to fourth embodiments of the present invention are different from the first to fourth embodiments in that the abnormal shift behavior is identified based on the speed and the acceleration in the step 4; the specific process is as follows:
step 4.1, acceleration data which are greater than or equal to 0 in all data acquired by I shifts are put into a set C + ，
Figure BDA0003980834580000044
Acceleration data less than 0 in all the data acquired by I shifts are put into a set C - ，
Figure BDA0003980834580000045
The maximum value of the shift running speed is denoted as v max ；
Wherein: v n For set C + The velocity of the nth data point in (c),
Figure BDA0003980834580000046
for set C + Acceleration v of nth data point in (b) m For set C - Speed of the mth data point in +.>
Figure BDA0003980834580000047
For set C - Acceleration of the mth data point;
step 4.2. Speed interval [0, v ] max ]Dividing the speed interval into a set Q with 1m/s as interval,
Figure BDA0003980834580000048
first, the
Figure BDA0003980834580000049
The intervals are denoted as Q σ First->
Figure BDA00039808345800000410
The midpoint of each interval is +.>
Figure BDA00039808345800000411
Wherein, the liquid crystal display device comprises a liquid crystal display device,
Figure BDA00039808345800000412
mathematical symbols rounded up;
interval Q σ Among the data acquisition points included are
Figure BDA00039808345800000413
The data acquisition points belong to set C + ，/>
Figure BDA00039808345800000414
The data acquisition points belong to set C - ；
Step 4.3. Section
Figure BDA00039808345800000415
Belongs to set C + And C - The data acquisition points of (1) are arranged from small to large according to the acceleration values, and respectively put into the collection +.>
Figure BDA00039808345800000416
And->
Figure BDA00039808345800000417
Step 4.4. Bring together
Figure BDA00039808345800000418
The 95 th quantile of the medium-speed value is denoted +.>
Figure BDA00039808345800000419
Set->
Figure BDA00039808345800000420
The 95 th quantile of the medium-speed value is denoted +.>
Figure BDA00039808345800000421
Figure BDA00039808345800000422
As a discrimination value of abnormal shift behavior at each speed section (here, the 95 th quantile of the acceleration value is taken as a threshold, namely, shift behavior which exceeds the acceleration value and is considered to be abnormal and biased, and then the acceleration threshold is given by fitting a polynomial to determine whether the abnormal shift behavior is the abnormal shift behavior), only behavior corresponding to 5% of acceleration is regarded as the biased shift behavior;
step 4.5. Differentiating the positive acceleration from the negative acceleration by the set of points expressed as P + 、P - As shown in the following formula;
respectively in the point set P + 、P - Drawing an acceleration threshold curve according to the data points in the model, wherein the abscissa is the speed, and the ordinate is the acceleration;
Figure BDA0003980834580000051
step 4.6. Respectively aligning the point sets P + 、P - Fitting a polynomial curve by using a least square method, taking the velocity v as an independent variable, the acceleration a as a dependent variable, and setting polynomial curve functions as a respectively + (v)、a - (v)；
Step 4.7. polynomial a + (v)、a - (v) Respectively with 2 to 5 th order (polynomial a + (v)、a - (v) The highest order M, H of (2) is respectively fitted by 2.3 and 4.5), and the fitting polynomial coefficient alpha is determined by taking the minimum of the sum of squares of deviation L (alpha) and L (beta) as the target 0 ,α 1 ,...,α M 、β 0 ,β 1 ,...,β H And a highest order M, H;
step 4.8. polynomial a after fitting + (v)、a - (v) Respectively used as a positive acceleration threshold curve and a negative acceleration threshold curve, and judging whether the operating passenger car has abnormal speed change behaviors according to the positive acceleration threshold curve and the negative acceleration threshold curve.
Other steps and parameters are the same as in one to four embodiments.
Specific embodiment six: the difference between the present embodiment and one to fifth embodiments is that the 95 th quantile determining method in the step 4.4 is as follows;
Figure BDA0003980834580000052
wherein:
Figure BDA0003980834580000053
representation set->
Figure BDA0003980834580000054
Middle->
Figure BDA0003980834580000055
Acceleration values of bits;
Figure BDA0003980834580000056
representation set->
Figure BDA0003980834580000057
Middle->
Figure BDA0003980834580000058
Acceleration values of bits.
Other steps and parameters are the same as in one of the first to fifth embodiments.
Seventh embodiment: this embodiment differs from one to six of the embodiments in that the respective pairs of points P in step 4.6 are + 、P - Fitting polynomial curves using least squaresTaking the velocity v as an independent variable and the acceleration a as an independent variable, and setting polynomial curve functions as a respectively + (v)、a - (v) As shown in the following formula;
Figure BDA0003980834580000061
wherein: m, H each represents a polynomial curve function a + (v)、a - (v) Is the highest order of (2); alpha 0 ,α 1 ,...,α M Representing the coefficients of the fitting polynomial, beta 0 ,β 1 ,...,β H Representing fitting polynomial coefficients; v M Represents the M-order of velocity v, v H Represents the H-order of velocity v, v j Represents the j-th order of velocity v, v l Representing the first order of velocity v.
Other steps and parameters are the same as in one of the first to sixth embodiments.
Eighth embodiment: this embodiment differs from one of the first to seventh embodiments in that in step 4.7, L (α) and L (β) are calculated as shown in the following formula;
Figure BDA0003980834580000062
other steps and parameters are the same as those of one of the first to seventh embodiments.
Detailed description nine: this embodiment differs from one to eight of the embodiments in that the fitted polynomial a in step 4.8 + (v)、a - (v) Respectively serving as a positive acceleration threshold curve and a negative acceleration threshold curve, and judging whether the operating passenger car has abnormal speed change behaviors according to the positive acceleration threshold curve and the negative acceleration threshold curve; the specific process is as follows:
step 4.8.1. K=1;
step 4.8.2 determining the acceleration of shift i after correction of the k data acquisition point coordinates
Figure BDA0003980834580000063
Corresponding speed->
Figure BDA0003980834580000064
Acceleration threshold value of->
Figure BDA0003980834580000065
And->
Figure BDA0003980834580000066
(i.e. a + (v)、a - (v) After the fitting is completed, each speed value in the acceleration threshold curve can correspond to one acceleration threshold value);
step 4.8.3. Judging
Figure BDA0003980834580000067
If so, go to step 4.8.4; otherwise, go to step 4.8.5;
step 4.8.4 judgment
Figure BDA0003980834580000068
If yes, judging that the operation passenger car has one abnormal acceleration behavior; otherwise, judging that the operation bus does not have abnormal acceleration behavior;
step 4.8.5 judge
Figure BDA0003980834580000069
If yes, judging that the operation passenger car has one abnormal deceleration action; otherwise, judging that the operation bus does not have abnormal deceleration behavior;
step 4.8.6. K=k+1, judging that K is less than or equal to K, if yes, turning to step 4.8.2; otherwise, go to step 4.8.7;
and 4.8.7, after traversing all accelerations, screening out all abnormal speed change behavior points.
Other steps and parameters are the same as in one to eight of the embodiments.
Detailed description ten: the present embodiment differs from one of the first to ninth embodiments in that the step 5 is to perform frequent abnormal shift section identification based on the abnormal shift behavior; the specific process is as follows:
the method comprises the steps of researching the spatial distribution situation of abnormal speed change behaviors, recording the occurrence position of the abnormal speed change behaviors as one-time abnormal speed change behavior data, and performing spatial clustering on the abnormal speed change behavior data by adopting a DBSCAN clustering method;
step 5.1, defining parameters; the specific process is as follows:
step 5.1.1. The coordinates of the position where the singular abnormal shift event occurs, r_p (x, y), are denoted as p ψ (x, y), the abnormal shift behavior data form a set D, the total being z, denoted as
Figure BDA0003980834580000071
Step 5.1.2. Calculating abnormal speed change behavior sample points with Euclidean distance
Figure BDA0003980834580000072
And->
Figure BDA0003980834580000073
Distance between
Figure BDA0003980834580000074
The formula is shown as follows;
Figure BDA0003980834580000075
wherein: γ, η ε [1, z ];
Figure BDA0003980834580000076
sample points +.>
Figure BDA0003980834580000077
The x-axis coordinate of the plane coordinate, m;
Figure BDA0003980834580000078
sample points +.>
Figure BDA0003980834580000079
The y-axis coordinate of the plane coordinate, m;
step 5.1.3 abnormal Shift Point of action p ψ Is represented as N pψ The neighborhood radius is epsilon, and the formula is shown as follows;
Figure BDA00039808345800000710
step 5.1.4. Point p ψ Neighborhood region
Figure BDA00039808345800000711
The minimum sample point in the range is denoted as lambda, point p ψ The conditions that need to be satisfied as core sample points are shown in the following formula;
Figure BDA00039808345800000712
wherein:
Figure BDA00039808345800000713
expressed as p ψ The number of sample points in the neighborhood of the core point;
step 5.1.5. Evaluating the clustering effect of the clustering samples by using the contour coefficients, and abnormal speed change behavior sample points
Figure BDA00039808345800000714
The contour coefficient calculation of (2) is shown in the following formula;
Figure BDA0003980834580000081
wherein:
Figure BDA0003980834580000082
for sample dot->
Figure BDA0003980834580000083
Average distance to other points within the cluster;
Figure BDA0003980834580000084
for sample dot->
Figure BDA0003980834580000085
Average distance to points in other clusters;
step 5.1.6. Use of overall contour coefficient average
Figure BDA0003980834580000086
As a sample overall clustering effect evaluation index, the specific calculation is shown in the following formula;
Figure BDA0003980834580000087
step 5.2, a specific clustering step; the specific process is as follows:
step 5.2.1, initializing a neighborhood radius epsilon, and according to the abnormal speed change behavior characteristic, not suggesting to select a larger epsilon, and taking epsilon=100deg.M;
step 5.2.2. Defining minimum sample points in the cluster according to road class, and the available minimum value is lambda 1 Maximum value lambda 2 (can flexibly adjust according to road grades, such as expressway, primary road and secondary road … …, can define different minimum sample point number value ranges), lambda is an integer, and the value range is [ lambda ] 1 ,λ 2 ]；
Step 5.2.3. Initializing, wherein the parameter h=0;
step 5.2.4.λ=λ 1 +h, input ε, λ;
5.2.5. adopting a DBSCAN clustering method to perform clustering on all abnormal speed change behavior points p in a set D ψ (x, y) completing spatial clustering;
step 5.2.6. Calculating the profile coefficients of the points of the sample of the abnormal speed-changing behaviors
Figure BDA0003980834580000088
Average value of overall profile coefficient
Figure BDA0003980834580000089
Step 5.2.7 determine lambda 1 +h≤λ 2 If yes, h=h+1, return to step 5.2.4; otherwise, enter step 5.2.8;
step 5.2.8. Calculating the overall profile factor maximum
Figure BDA00039808345800000810
The formula is shown as follows;
Figure BDA00039808345800000811
each lambda is used as input to correspondingly calculate an average value of the overall profile coefficients, and the step 5.2.8 is to take the maximum value of the overall profile coefficients;
step 5.2.9 determining
Figure BDA00039808345800000812
The corresponding parameter lambda is lambda 0 I.e. final critical parameter λ=λ 0 ；
Step 5.3. Determining the clustering parameter ε=100deg.M, λ=λ of DBSCAN 0 Is the optimal cluster;
step 5.4. Point of all abnormal Shift behavior p in set D ψ (x, y) outputting the number of clusters and the number of sample points in each cluster after optimal clustering, wherein the number of clusters is the number of frequent abnormal speed change road sections, and the number of sample points in each cluster is the number of abnormal speed change behaviors of the road sections;
and 5.5. Sample points are distributed on the road network, so that the clustered cluster shape (the clustered cluster shape is a road section because all the sample points are on the road network) is attached to the road network, the obtained cluster is a frequently-occurring abnormal speed change road section, and the edge points of the cluster along the road network direction are the starting and ending points of the frequently-occurring abnormal speed change road section.
Other steps and parameters are the same as in one of the first to ninth embodiments.
The present invention is capable of other and further embodiments and its several details are capable of modification and variation in light of the present invention, as will be apparent to those skilled in the art, without departing from the spirit and scope of the invention as defined in the appended claims.

Claims (10)

1. A method for identifying abnormal speed change behavior frequent road sections of a commercial passenger car is characterized by comprising the following steps of: the method comprises the following specific processes:
step 1, basic data acquisition;
step 2, carrying out coordinate deviation correction based on the basic data;
step 3, correcting and calculating the speed and the acceleration based on the basic data and the coordinates;
step 4, identifying abnormal speed change behaviors based on the speed and the acceleration;
and 5, carrying out frequent abnormal speed change road section identification based on the abnormal speed change behavior.
2. The method for identifying abnormal speed change behavior frequent road segments of a commercial passenger car according to claim 1, wherein the method comprises the following steps: the basic data is acquired in the step 1; the specific process is as follows:
step 1.1, extracting GPS data of all operating shifts in the past continuous N days of a commercial bus line, wherein the GPS data comprises acquisition time, sampling interval, longitude and latitude, vehicle ID and shift direction information;
step 1.2, the space distribution of the frequently-occurring road sections of the abnormal speed change behavior in the up-down direction of the operating passenger car has obvious difference, and the up-down direction operating road of the operating passenger car is selected as a research object, so that the method is also applicable to the down-direction operating road;
step 1.3, defining an operation shift set of the uplink direction of the line as i= { I: i=1, 2, 3., I }; shift i has K GPS data acquisition points in total, and the K data acquisition point coordinates of shift i are expressed as
Figure FDA0003980834570000011
Wherein lng and lat respectively represent longitude and latitude, and the unit is degree; the GPS sampling interval is expressed as T, and the unit is s; k=1, 2,3,;
and 1.4, extracting a GIS road network of a running line passing region of the operating passenger car, wherein the GIS road network comprises a road name, a road number and road grade information.
3. The method for identifying abnormal speed change behavior frequent road segments of a commercial passenger car according to claim 2, wherein the method comprises the following steps: in the step 2, coordinate deviation correction is performed based on basic data; the specific process is as follows:
2.1, adopting a map matching algorithm based on a hidden Markov model to finish map matching of the GPS data acquisition points and the GIS road network, wherein the matched road network is the road network where the GPS data acquisition points are actually located;
step 2.2. The coordinates of the GPS data acquisition points after map matching are the actual positions of the GPS data acquisition points on the road network, and are expressed as
Figure FDA0003980834570000012
Step 2.3. Matching the coordinates of the map
Figure FDA0003980834570000013
Unified conversion into planar coordinates->
Figure FDA0003980834570000014
And x and y are projection distances from the GPS data acquisition point to the central meridian and the equator in a coordinate system taking the intersection point of the equator and the central meridian as the origin of coordinates, wherein the unit is m.
4. The method for identifying abnormal speed change behavior frequent road segments of a commercial passenger car according to claim 3, wherein the method comprises the following steps: in the step 3, the speed and the acceleration are calculated based on basic data and coordinate deviation correction; the specific process is as follows:
step 3.1, the speed after correction of the k data acquisition point coordinates of shift i is expressed as
Figure FDA0003980834570000021
The unit is thatm/s; calculation speed->
Figure FDA0003980834570000022
The formula is shown as follows;
Figure FDA0003980834570000023
wherein:
Figure FDA0003980834570000024
the x-axis coordinates of plane coordinates after correction of the k+1 data acquisition point coordinates and the k data acquisition point coordinates of shift i are respectively given in m;
Figure FDA0003980834570000025
the y-axis coordinates of plane coordinates after correction of the k+1 data acquisition point coordinates and the k data acquisition point coordinates of shift i are respectively given in m;
step 3.2, the acceleration after correction of the k data acquisition point coordinates of shift i is expressed as
Figure FDA0003980834570000026
In m/s 2 The method comprises the steps of carrying out a first treatment on the surface of the Calculate acceleration->
Figure FDA0003980834570000027
The formula is shown as follows;
Figure FDA0003980834570000028
wherein:
Figure FDA0003980834570000029
the speed after correction is obtained for the k-1 data acquisition point coordinates of shift i.
5. The method for identifying abnormal speed change behavior frequent road segments of a commercial passenger car according to claim 4, wherein the method comprises the following steps: in the step 4, abnormal speed change behavior identification is carried out based on the speed and the acceleration; the specific process is as follows:
step 4.1, acceleration data which are greater than or equal to 0 in all data acquired by I shifts are put into a set C + ，
Figure FDA00039808345700000210
Acceleration data less than 0 in all the data acquired by I shifts are put into a set C - ，
Figure FDA00039808345700000211
The maximum value of the shift running speed is denoted as v max ；
Wherein: v n For set C + The velocity of the nth data point in (c),
Figure FDA00039808345700000212
for set C + Acceleration v of nth data point in (b) m For set C - Speed of the mth data point in +.>
Figure FDA00039808345700000213
For set C - Acceleration of the mth data point;
step 4.2. Speed interval [0, v ] max ]Dividing the speed interval into a set Q with 1m/s as interval,
Figure FDA00039808345700000214
first, the
Figure FDA00039808345700000215
The intervals are denoted as Q σ First->
Figure FDA00039808345700000216
The midpoint of each interval is +.>
Figure FDA00039808345700000217
Wherein, the liquid crystal display device comprises a liquid crystal display device,
Figure FDA0003980834570000031
mathematical symbols rounded up;
interval Q σ Among the data acquisition points included are
Figure FDA0003980834570000032
The data acquisition points belong to set C + ，/>
Figure FDA0003980834570000033
The data acquisition points belong to set C - ；
Step 4.3. Section
Figure FDA0003980834570000034
Belongs to set C + And C - The data acquisition points of (1) are arranged from small to large according to the acceleration values, and respectively put into the collection +.>
Figure FDA0003980834570000035
And->
Figure FDA0003980834570000036
Step 4.4. Bring together
Figure FDA0003980834570000037
The 95 th quantile of the medium-speed value is denoted +.>
Figure FDA0003980834570000038
Set->
Figure FDA0003980834570000039
The 95 th quantile of the medium-speed value is denoted +.>
Figure FDA00039808345700000310
Figure FDA00039808345700000311
As a distinguishing value of abnormal speed change behaviors in each speed interval, only behaviors corresponding to 5% of acceleration are regarded as deflection excitation speed change behaviors;
step 4.5. Differentiating the positive acceleration from the negative acceleration by the set of points expressed as P + 、P - As shown in the following formula;
respectively in the point set P + 、P - Drawing an acceleration threshold curve according to the data points in the model, wherein the abscissa is the speed, and the ordinate is the acceleration;
Figure FDA00039808345700000312
step 4.6. Respectively aligning the point sets P + 、P - Fitting a polynomial curve by using a least square method, taking the velocity v as an independent variable, the acceleration a as a dependent variable, and setting polynomial curve functions as a respectively + (v)、a - (v)；
Step 4.7. polynomial a + (v)、a - (v) Fitting with 2-5 steps respectively, and determining fitting polynomial coefficient alpha by taking the sum of squares of deviation L (alpha) and L (beta) as the minimum target 0 ,α 1 ,...,α M 、β 0 ,β 1 ,...,β H And a highest order M, H;
step 4.8. polynomial a after fitting + (v)、a - (v) Respectively used as a positive acceleration threshold curve and a negative acceleration threshold curve, and judging whether the operating passenger car has abnormal speed change behaviors according to the positive acceleration threshold curve and the negative acceleration threshold curve.
6. The method for identifying abnormal speed change behavior frequent road segments of a commercial passenger car according to claim 5, wherein the method comprises the following steps: the 95 th quantile determining method in the step 4.4 is shown in the following formula;
Figure FDA00039808345700000313
wherein:
Figure FDA00039808345700000314
representation set->
Figure FDA00039808345700000315
Middle->
Figure FDA00039808345700000316
Acceleration values of bits;
Figure FDA0003980834570000041
representation set->
Figure FDA0003980834570000042
Middle->
Figure FDA0003980834570000043
Acceleration values of bits.
7. The method for identifying abnormal speed change behavior frequent road segments of a commercial passenger car according to claim 6, wherein the method comprises the following steps: the corresponding point sets P in the step 4.6 + 、P - Fitting a polynomial curve by using a least square method, taking the velocity v as an independent variable, the acceleration a as a dependent variable, and setting polynomial curve functions as a respectively + (v)、a - (v) As shown in the following formula;
Figure FDA0003980834570000044
wherein: m, H each represents a polynomial curve function a + (v)、a - (v) Is the highest order of (2); alpha 0 ,α 1 ,...,α M Representation ofFitting polynomial coefficients, beta 0 ,β 1 ,...,β H Representing fitting polynomial coefficients; v M Represents the M-order of velocity v, v H Represents the H-order of velocity v, v j Represents the j-th order of velocity v, v l Representing the first order of velocity v.
8. The method for identifying abnormal speed change behavior frequent road segments of a commercial passenger car according to claim 7, wherein the method comprises the following steps: in the step 4.7, L (alpha) and L (beta) are calculated as shown in the following formula;
Figure FDA0003980834570000045
9. the method for identifying abnormal speed change behavior frequent road segments of a commercial passenger car according to claim 8, wherein the method comprises the following steps: the polynomial a after fitting in the step 4.8 + (v)、a - (v) Respectively serving as a positive acceleration threshold curve and a negative acceleration threshold curve, and judging whether the operating passenger car has abnormal speed change behaviors according to the positive acceleration threshold curve and the negative acceleration threshold curve; the specific process is as follows:
step 4.8.1. K=1;
step 4.8.2 determining the acceleration of shift i after correction of the k data acquisition point coordinates
Figure FDA0003980834570000046
Corresponding speed->
Figure FDA0003980834570000047
Acceleration threshold value of->
Figure FDA0003980834570000048
And->
Figure FDA0003980834570000049
Step 4.8.3. Judging
Figure FDA00039808345700000410
If so, go to step 4.8.4; otherwise, go to step 4.8.5;
step 4.8.4 judgment
Figure FDA00039808345700000411
If yes, judging that the operation passenger car has one abnormal acceleration behavior; otherwise, judging that the operation bus does not have abnormal acceleration behavior;
step 4.8.5 judge
Figure FDA00039808345700000412
If yes, judging that the operation passenger car has one abnormal deceleration action; otherwise, judging that the operation bus does not have abnormal deceleration behavior;
step 4.8.6. K=k+1, judging that K is less than or equal to K, if yes, turning to step 4.8.2; otherwise, go to step 4.8.7;
and 4.8.7, after traversing all accelerations, screening out all abnormal speed change behavior points.
10. The method for identifying abnormal speed change behavior frequent road segments of a commercial passenger car according to claim 9, wherein the method comprises the following steps: in the step 5, the identification of frequent abnormal speed change road sections is carried out based on abnormal speed change behaviors; the specific process is as follows:
step 5.1, defining parameters;
the specific process is as follows:
step 5.1.1. The coordinates of the position where the singular abnormal shift event occurs, r_p (x, y), are denoted as p ψ (x, y), the abnormal shift behavior data form a set D, the total being z, denoted as
Figure FDA0003980834570000051
Step 5.1.2. Calculating abnormal speed change behavior sample points with Euclidean distance
Figure FDA0003980834570000052
And->
Figure FDA0003980834570000053
Distance between->
Figure FDA0003980834570000054
The formula is shown as follows;
Figure FDA0003980834570000055
wherein: γ, η ε [1, z ];
Figure FDA0003980834570000056
sample points +.>
Figure FDA0003980834570000057
The x-axis coordinate of the plane coordinate, m;
Figure FDA0003980834570000058
sample points +.>
Figure FDA0003980834570000059
The y-axis coordinate of the plane coordinate, m;
step 5.1.3 abnormal Shift Point of action p ψ The neighborhood of (1) is expressed as
Figure FDA00039808345700000510
The neighborhood radius is epsilon, and the formula is shown as follows;
Figure FDA00039808345700000511
step 5.1.4. Point p ψ Neighborhood region
Figure FDA00039808345700000512
The minimum sample point in the range is expressed as lambda, pointp ψ The conditions that need to be satisfied as core sample points are shown in the following formula;
Figure FDA00039808345700000513
wherein:
Figure FDA00039808345700000514
expressed as p ψ The number of sample points in the neighborhood of the core point;
step 5.1.5. Evaluating the clustering effect of the clustering samples by using the contour coefficients, and abnormal speed change behavior sample points
Figure FDA00039808345700000515
The contour coefficient calculation of (2) is shown in the following formula;
Figure FDA0003980834570000061
wherein:
Figure FDA0003980834570000062
for sample dot->
Figure FDA0003980834570000063
Average distance to other points within the cluster;
Figure FDA0003980834570000064
for sample dot->
Figure FDA0003980834570000065
Average distance to points in other clusters;
step 5.1.6. Use of overall contour coefficient average
Figure FDA0003980834570000066
As a sample overall clustering effect evaluation index, the specific calculation is shown in the following formula;
Figure FDA0003980834570000067
step 5.2, a specific clustering step;
the specific process is as follows:
step 5.2.1, initializing a neighborhood radius epsilon, and taking epsilon=100deg.M;
step 5.2.2. Defining minimum sample points in the cluster according to road class, and the available minimum value is lambda 1 Maximum value lambda 2 Lambda is an integer, and the value range is [ lambda ] 1 ,λ 2 ]；
Step 5.2.3. Initializing, wherein the parameter h=0;
step 5.2.4.λ=λ 1 +h, input ε, λ;
step 5.2.5. Adopting DBSCAN clustering method to make all abnormal speed change action points p in the set D ψ (x, y) completing spatial clustering;
step 5.2.6. Calculating the profile coefficients of the points of the sample of the abnormal speed-changing behaviors
Figure FDA0003980834570000068
Mean value of overall profile coefficients->
Figure FDA0003980834570000069
Step 5.2.7 determine lambda 1 +h≤λ 2 If yes, h=h+1, return to step 5.2.4; otherwise, enter step 5.2.8;
step 5.2.8. Calculating the overall profile factor maximum
Figure FDA00039808345700000610
The formula is shown as follows;
Figure FDA00039808345700000611
step 5.2.9 determining
Figure FDA00039808345700000612
The corresponding parameter lambda is lambda 0 I.e. final critical parameter λ=λ 0 ；
Step 5.3. Determining the clustering parameter ε=100deg.M, λ=λ of DBSCAN 0 Is the optimal cluster;
step 5.4. Point of all abnormal Shift behavior p in set D ψ (x, y) outputting the number of clusters and the number of sample points in each cluster after optimal clustering, wherein the number of clusters is the number of frequent abnormal speed change road sections, and the number of sample points in each cluster is the number of abnormal speed change behaviors of the road sections;
and 5.5. Sample points are distributed on the road network, so that the clustered cluster shape is attached to the road network, the obtained cluster is a frequently-occurring abnormal speed change road section, and the edge point of the cluster along the road network direction is the starting and ending point of the frequently-occurring abnormal speed change road section.
CN202211578800.2A 2022-12-05 2022-12-05 Method for identifying abnormal speed change behavior frequent road sections of operating passenger car Active CN116383678B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202211578800.2A CN116383678B (en) 2022-12-05 2022-12-05 Method for identifying abnormal speed change behavior frequent road sections of operating passenger car

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202211578800.2A CN116383678B (en) 2022-12-05 2022-12-05 Method for identifying abnormal speed change behavior frequent road sections of operating passenger car

Publications (2)

Publication Number Publication Date
CN116383678A true CN116383678A (en) 2023-07-04
CN116383678B CN116383678B (en) 2024-05-03

Family

ID=86962056

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202211578800.2A Active CN116383678B (en) 2022-12-05 2022-12-05 Method for identifying abnormal speed change behavior frequent road sections of operating passenger car

Country Status (1)

Country Link
CN (1) CN116383678B (en)

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20110029184A1 (en) * 2009-07-31 2011-02-03 Systems and Advances Technologies Engineering S.r.I. (S.A.T.E.) Road Vehicle Drive Behaviour Analysis Method
CN104933863A (en) * 2015-06-02 2015-09-23 福建工程学院 Method and system for recognizing abnormal segment of traffic road
CN109059939A (en) * 2018-06-27 2018-12-21 湖南智慧畅行交通科技有限公司 Map-matching algorithm based on Hidden Markov Model
CN110239559A (en) * 2019-07-02 2019-09-17 绍兴数鸿科技有限公司 Dangerous driving vehicle checking method and device based on new energy car data
CN114756599A (en) * 2022-04-06 2022-07-15 扬州大学 Driver abnormal speed change identification method based on vehicle GPS data

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20110029184A1 (en) * 2009-07-31 2011-02-03 Systems and Advances Technologies Engineering S.r.I. (S.A.T.E.) Road Vehicle Drive Behaviour Analysis Method
CN104933863A (en) * 2015-06-02 2015-09-23 福建工程学院 Method and system for recognizing abnormal segment of traffic road
CN109059939A (en) * 2018-06-27 2018-12-21 湖南智慧畅行交通科技有限公司 Map-matching algorithm based on Hidden Markov Model
CN110239559A (en) * 2019-07-02 2019-09-17 绍兴数鸿科技有限公司 Dangerous driving vehicle checking method and device based on new energy car data
CN114756599A (en) * 2022-04-06 2022-07-15 扬州大学 Driver abnormal speed change identification method based on vehicle GPS data

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
张迪: "基于浮动车GPS数据分析的车辆相对异常驾驶行为研究", 《中国优秀硕士学位论文全文数据库 (工程科技Ⅱ辑)》, no. 2019, 15 January 2019 (2019-01-15) *

Also Published As

Publication number Publication date
CN116383678B (en) 2024-05-03

Similar Documents

Publication Publication Date Title
EP2710572B1 (en) Vehicle data analysis method and vehicle data analysis system
CN109885040B (en) Vehicle driving control right distribution system in man-machine driving
CN110428621B (en) Track data-based monitoring and early warning method for dangerous driving behavior of floating car
CN109094482B (en) Vehicle operation risk assessment information acquisition system, risk assessment system and method
CN106956680B (en) Electric automobile driving behavior recognition analysis method
EP3578433B1 (en) Method for estimating an accident risk of an autonomous vehicle
CN105975757A (en) Urgent speed reduction behavior recognition method based on vehicle driving data
CN113095387B (en) Road risk identification method based on networking vehicle-mounted ADAS
CN112356849A (en) Early warning method and device for dangerous driving
CN114912689A (en) Map grid index and XGBOST-based over-limit vehicle destination prediction method and system
CN114999181A (en) ETC system data-based highway vehicle speed abnormity identification method
CN114379559A (en) Driving risk evaluation feature sketch method based on vehicle information acquisition system
CN116383678B (en) Method for identifying abnormal speed change behavior frequent road sections of operating passenger car
CN111775948B (en) Driving behavior analysis method and device
CN115909752A (en) Sharp turn recognition and statistics method based on historical data of vehicle user
CN114119256A (en) UBI dangerous chemical vehicle driving behavior acquisition and analysis system and premium discount method
CN112036709B (en) Random forest based rainfall weather expressway secondary accident cause analysis method
CN110544378B (en) Method for judging traffic jam condition of mobile phone user
CN114255590B (en) Traffic operation analysis method based on RFID data
CN112530164B (en) Active driving safety early warning method and system for high-altitude area road
CN116665456B (en) Method for evaluating traffic state by combining high-dimensional index dimension reduction processing
CN117194930B (en) Tunnel road section safety monitoring method based on vehicle driving data
CN114944083B (en) Method for judging distance between running vehicle on expressway and front vehicle
CN115966100B (en) Driving safety control method and system
CN116343484B (en) Traffic accident identification method, terminal and storage medium

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