CN108682153B - Urban road traffic jam state discrimination method based on RFID electronic license plate data - Google Patents

Urban road traffic jam state discrimination method based on RFID electronic license plate data Download PDF

Info

Publication number
CN108682153B
CN108682153B CN201810549401.0A CN201810549401A CN108682153B CN 108682153 B CN108682153 B CN 108682153B CN 201810549401 A CN201810549401 A CN 201810549401A CN 108682153 B CN108682153 B CN 108682153B
Authority
CN
China
Prior art keywords
clustering
traffic jam
center
points
centers
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
CN201810549401.0A
Other languages
Chinese (zh)
Other versions
CN108682153A (en
Inventor
郑林江
刘卫宁
刘亚东
孙棣华
赵敏
陈莉
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.)
Chongqing University
Original Assignee
Chongqing 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 Chongqing University filed Critical Chongqing University
Priority to CN201810549401.0A priority Critical patent/CN108682153B/en
Publication of CN108682153A publication Critical patent/CN108682153A/en
Application granted granted Critical
Publication of CN108682153B publication Critical patent/CN108682153B/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
    • G08G1/0108—Measuring and analyzing of parameters relative to traffic conditions based on the source of data
    • G08G1/0112—Measuring and analyzing of parameters relative to traffic conditions based on the source of data from the vehicle, e.g. floating car data [FCD]
    • G—PHYSICS
    • G08—SIGNALLING
    • G08G—TRAFFIC CONTROL SYSTEMS
    • G08G1/00—Traffic control systems for road vehicles
    • G08G1/01—Detecting movement of traffic to be counted or controlled
    • G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
    • G08G1/0125—Traffic data processing
    • G08G1/0133—Traffic data processing for classifying traffic situation
    • 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
    • G08G1/017—Detecting movement of traffic to be counted or controlled identifying vehicles

Landscapes

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

Abstract

The invention discloses an urban road traffic jam state discrimination method based on RFID electronic license plate data, which comprises the following steps: 1) acquiring traffic flow evaluation parameters including equivalent number and average speed of standard vehicles through RFID electronic license plate data; 2) taking the equivalent number and the average speed of the standard vehicles obtained in the step 1) as two dimensions of clustering operation, and performing clustering operation to obtain the optimal clustering number and clustering result; 3) and sorting the cluster centers from large to small according to the values of the vertical coordinate projection points, projecting the cluster centers with the same values of the vertical coordinate projection points to the horizontal coordinate, and sorting the cluster centers from small to large according to the values of the horizontal coordinate projection points, wherein the cluster centers corresponding to the sorted projection points represent the continuous change trend of the traffic jam state from unblocked to jammed. According to the invention, the number of equivalents of the standard vehicles is used for replacing the traffic volume as the dimension of the GEFCM clustering algorithm, so that the traffic jam state can be better reflected, and compared with the traditional fuzzy C mean value algorithm, the GEFCM is used for avoiding the influence on the final clustering result due to the difference of sample class capacity, so that the clustering accuracy is improved; and the problem of matching between the clustering center and the specific traffic jam state is solved.

Description

Urban road traffic jam state discrimination method based on RFID electronic license plate data
Technical Field
The invention relates to the technical field of traffic data processing, in particular to a method for judging urban road traffic jam state based on RFID electronic license plate data.
Background
The working principle of the RFID technology, namely, the Radio Frequency Identification (Radio Frequency Identification) technology, is to implement a self-Identification function by using an Identification technology of non-contact bidirectional communication in a Radio Frequency manner. The RFID technology has the advantages of high confidentiality, long reading and writing distance, capability of identifying objects moving at high speed such as automobiles and the like, non-contact bidirectional communication and the like, and can work in severe environment. By combining the RFID technology with the communication technology, the Internet technology and the like, the RFID technology is widely applied to the fields of Internet of things, intelligent transportation, anti-counterfeiting of commodities and the like at present.
With the development of social economy, the quantity of motor vehicles in China continuously rises, and people go out more frequently, which inevitably brings certain pressure to road traffic, so that traffic jam is caused.
At present, the urban road traffic jam state judgment is mainly based on an automatic judgment technology, and with the continuous development and improvement of technical theories in the traffic field, technologies such as a fuzzy theory, a neural network, a cross-discipline theory and the like are increasingly applied to the research of traffic jam judgment. The traffic jam state judgment is based on reasonable selection of traffic flow evaluation parameters, and based on different types of sample data, the types, the difficulty degree and the accuracy of the selection of the traffic flow evaluation parameters are different. The RFID electronic license plate data is based on the RFID electronic license plate data, and the advantages of the RFID technology in the traffic field are embodied in the aspects of fast vehicle identification technology, no influence of weather conditions on vehicle identification, comprehensive vehicle identification information and the like, so the RFID electronic license plate data is more suitable for serving as basic data for judging traffic jam compared with other traffic flow data. The existing traffic jam state judging technology is usually based on quantitative analysis of various traffic flow parameters, but the traffic jam states of different roads are very different and are different even at the same time point, so that a more universal method is needed for judging the traffic jam state.
Disclosure of Invention
In view of the above, the present invention provides a more general traffic jam state determination method based on RFID electronic license plate data.
The purpose of the invention is realized by the following technical scheme: a city road section travel time prediction method based on RFID data comprises the following steps:
1) acquiring traffic flow evaluation parameters including equivalent number and average speed of standard vehicles through RFID electronic license plate data;
2) taking the equivalent number and the average speed of the standard vehicles obtained in the step 1) as two dimensions of clustering operation, and performing clustering operation to obtain the optimal clustering number and clustering result;
3) taking the average speed as a vertical coordinate, taking the equivalent number of standard vehicles as a horizontal coordinate, sequencing the clustering centers from large to small according to the values of vertical coordinate projection points, projecting the clustering center points with the same values of the vertical coordinate projection points to the horizontal coordinate, sequencing according to the values of the horizontal coordinate projection points from small to large, representing the continuous change trend of the traffic jam state from unblocked to jammed by the clustering center points corresponding to the sequenced projection points, and then matching the traffic jam state of the clustering centers according to the leftmost state continuous matching principle.
Further, the step 1) specifically comprises the following steps:
11) the RFID acquisition points acquire RFID electronic license plate data to obtain the traffic volume of all vehicles and the traffic volume of large vehicles in a time period, and the equivalent number of standard vehicles in the time period is calculated according to the following formula:
pcu＝n+n2；
wherein pcu represents the number of equivalents of standard vehicle, n represents the traffic volume of all vehicle types, n2Representing the traffic volume of a large vehicle;
12) the average speed of the vehicle is calculated by:
Figure BDA0001680752890000021
in the formula, v represents the average speed in km/h, L represents the distance between two acquisition points of the road section in meters, ti1And ti2Respectively representing the time of the ith vehicle passing through the first RFID acquisition point and the second RFID acquisition point, wherein the unit is second, and n represents the traffic volume of all vehicle types in the time period.
Further, the clustering algorithm of step 2) specifically includes the following steps:
21) let the number c of cluster centers be 2;
22) initializing membership matrix uijAt initialization time uijIs a value of [0,1]A random number within a range;
23) calculating each clustering center c according to the membership matrixiThe values in each dimension:
Figure BDA0001680752890000022
in the formula, m is a fuzzy weighting coefficient, n represents the data size of sample data, uijMembership, u, for the jth sample belonging to the ith cluster centerikMembership, x, for the kth sample belonging to the ith cluster centerjRepresenting jth sample data;
24) according to the cluster center ciCalculating a membership matrix uij：
Figure BDA0001680752890000031
In the formula, m is a fuzzy weighting coefficient, c is the number of clustering centers, ciDenotes the ith cluster center, ckDenotes the k-th cluster center, xjDenotes the jth sample data, uijMembership, u, for the jth sample belonging to the ith cluster centerisMembership, u, of the s-th sample to the i-th cluster centerrsMembership, u, of the s-th sample to the r-th cluster centerrjThe degree of membership of the jth sample belonging to the r-th cluster center. (ii) a
25) If the objective function value is larger than the target function value, returning to the step iii, otherwise, performing the step vi;
26) at this time, the clustering operation under the clustering center number c is completed, let c be c +1, if c be less than or equal to 2ln n, the adjacency matrix expression form of the clustering result under the current clustering center number is calculated, and the adjacency matrix meaning is as follows: if the element of the ith row and the jth column is 1, the ith sample data and the jth sample data belong to the same category, and the step ii is returned after the calculation of the adjacency matrix is completed; if c is more than 2ln, go to step vii;
27) at this time, the clustering operation under all the clustering center numbers is completed, the adjacent matrixes in the step vi of the clustering operation under each clustering center number are counted, all the adjacent matrixes are added to obtain a total adjacent matrix OJ(ii) a Utilizing TreeMap to count the number of the clustering centers and the corresponding iteration times, and enabling the key of TreeMap to be the number of the clustering centers and the value to be the iteration times; mixing O withJSubtracting 1 from the value of not 0, and counting the current OJThe number num of the middle connected subgraphs is judged, whether a key with num exists in the TreeMap is judged, if not, the num is added into the TreeMap, and the corresponding value is 1; if a num key already exists in TreeMap, the value of the key is increased by 1 until OJAll elements in the formula are 0;
28) selecting a key with the largest TreeMap median as the number of clustering centers, and taking the clustering result of the number of the clustering centers as the clustering result of the sample data; if keys with the same value exist, judging by using a clustering effect evaluation index, respectively calculating the clustering effect evaluation index value under the condition that the keys with the same value are taken as the number of clustering centers, selecting the number of clustering centers with a smaller value of the clustering effect evaluation index as the number of selected clustering centers, wherein the calculation formula of the clustering effect evaluation index is as follows:
Figure BDA0001680752890000032
in the formula, c is the number of clustering centers, m is a fuzzy weighting coefficient, n represents the data volume of sample data, and uijMembership of the jth sample to the ith cluster center, ciDenotes the ith cluster center, ckDenotes the k-th cluster center, xjThe jth sample data is represented.
Further, the step 3) specifically comprises the following steps:
31) drawing the clustering center points on coordinate axes which respectively take the equivalent number of the standard vehicle and the average speed as horizontal and vertical coordinates;
32) sorting the clustering centers from large to small according to the values of the vertical coordinate projection points;
33) if the same vertical coordinate projection values exist, for the clustering center points with the same vertical coordinate projection values, the clustering center points are projected to the horizontal coordinate and are sorted from small to large according to the horizontal coordinate projection values, and the clustering center points corresponding to the sorted projection points represent the continuous change trend of the traffic jam state from unblocked to jammed;
34) identifying the traffic jam state of the clustering center according to a leftmost state continuous matching principle; the leftmost state continuous matching principle is as follows: c sequenced clustering centers are always matched with the front c traffic jam states TC one by one from left to right; the traffic jam state TC is divided into 6 types of very smooth, unblocked, slow-moving, light jam, jam and serious jam which respectively correspond to the traffic jam state TC<TC1,TC2,…,TC6>；
35) And obtaining the traffic jam states of all sample points around all the clustering centers according to the traffic jam states of the clustering centers.
Due to the adoption of the technical scheme, the invention has the following advantages:
the method has the advantages that the number of equivalents of the standard vehicles is used for replacing the traffic volume to serve as the dimension of the GEFCM clustering algorithm, the traffic jam state can be better reflected, compared with the traditional fuzzy C mean value algorithm, the influence of different sample class capacities on the final clustering result can be avoided by applying the GEFCM, and the clustering accuracy is improved.
Secondly, a 'sorting rule of projection points of the clustering center' and a 'continuous matching of the leftmost state' principle are provided, so that the problem of matching between the clustering center and the specific traffic jam state is solved.
Additional advantages, objects, and features of the invention will be set forth in part in the description which follows and in part will become apparent to those having ordinary skill in the art upon examination of the following or may be learned from practice of the invention. The objectives and other advantages of the invention may be realized and attained by the means of the instrumentalities and combinations particularly pointed out hereinafter.
Drawings
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention will be further described in detail with reference to the accompanying drawings:
FIG. 1 is a flow chart of the present invention;
FIG. 2 is a schematic diagram of a projection point of a cluster center to a vertical coordinate;
FIG. 3 is a schematic diagram of cluster center points with the same ordinate projection value;
fig. 4 is an example of a continuous change process of the traffic congestion state.
Detailed Description
The embodiments of the present invention are described below with reference to specific embodiments, and other advantages and effects of the present invention will be easily understood by those skilled in the art from the disclosure of the present specification. The invention is capable of other and different embodiments and of being practiced or of being carried out in various ways, and its several details are capable of modification in various respects, all without departing from the spirit and scope of the present invention.
Referring to fig. 1, the invention provides a method for judging urban road traffic jam state based on RFID electronic license plate data, comprising the following steps:
1) and calculating a traffic flow evaluation parameter, wherein the evaluation parameter refers to the equivalent number and the average speed of the road section standard vehicles in the time period. The time period is divided by a time threshold value T before the traffic flow evaluation parameter is calculated, T selected by the invention is 10 minutes, and sample data of each day can be divided into 144 equal parts, namely the time period is divided into [00:00,00:10 ], [00:10,00:20 ], [00:20,00:30 ], …, [23:50,00: 00). The standard vehicle equivalent number refers to the traffic volume number of various motor vehicles and non-motor vehicles running on an urban road, which is converted into a standard vehicle type according to a certain vehicle type conversion coefficient. The following is a process of calculating traffic flow evaluation parameters:
i) before the number of the equivalent of the standard vehicle is calculated, the traffic volume is calculated. The traffic volume is the number of all types of vehicles running on the road, and when the traffic volume is counted, the situation that the data of two RFID acquisition points in a road section are not matched can occur, which can be related to the deployment position of an RFID acquisition point device, and the running situations of the vehicles are divided into the following 3 situations due to the overlong distance between the two RFID acquisition points and the occurrence of other road sections between the two RFID acquisition points:
the vehicle passes through the acquisition point 1 and the acquisition point 2 at the same time.
② the vehicle passes through the collection point 1 but does not pass through the collection point 2.
③ the vehicle does not pass the acquisition point 1, but passes the acquisition point 2.
In view of the three situations, the data of the collection point with larger statistical traffic volume is selected in each time period as the traffic volume in the time period, and the traffic volume of the large-scale vehicle is calculated in the same way. The input tuple for calculating the traffic volume obtained from the original RFID electronic license plate data is as follows:
＜rfid1,passtime1,rfid2,passtime2,carType＞；
wherein rfid1Is the first RFID acquisition Point, RFID, passed2Is the second RFID acquisition Point, past1Is the time of passage through the first RFID acquisition Point, past2Is the time of passing the first RFID acquisition point, the type of the vehicle is the type of the vehicle, and a tuple can describe the traffic situation of a vehicle.
According to the input tuples, the traffic volume of all vehicles and the traffic volume of large vehicles in each time period are obtained, and the number of the equivalent of the standard vehicles in each time period is calculated according to the following formula:
pcu＝n+n2；
n in the formula represents the traffic volume of all vehicle types, n2Representing the amount of traffic of a large vehicle.
The resulting output tuple is as follows:
＜Ti,pcui＞；
wherein T isiDenotes the ith time period, pcuiIndicating the number of standard vehicle equivalents in the ith time period.
ii) the calculation of the average speed only needs to acquire the specific time when the vehicle passes through two acquisition points in a certain time period, and the input tuple for calculating the average speed obtained by the original RFID electronic license plate data is as follows:
＜rfid1,passtime1,rfid2,passtime2,L＞；
wherein rfid1Is the first RFID acquisition Point, RFID, passed2Is the second RFID acquisition Point, past1Is the time of passage through the first RFID acquisition Point, past2Is the time to pass the second RFID acquisition point and L is the distance between the two RFID acquisition points.
The average speed of all vehicles over a certain time period is then calculated according to the following formula:
Figure BDA0001680752890000061
in the formula, v represents average speed (km/h), L represents distance between two acquisition points of the road section (meter), and ti1And ti2Respectively, the time (in seconds) for the ith vehicle to pass through the first RFID acquisition point and the second RFID acquisition point, and n represents the traffic volume (note not the number of standard vehicle equivalents) in the time period. It should be noted here that 3.6 in the formula is to convert the unit from m/s to km/h.
The resulting output tuple is as follows:
＜Ti,vi＞
wherein T isiDenotes the i-th time period, viRepresenting the average speed of all vehicles during the ith time period
2) And (4) taking the equivalent number of the standard vehicles and the average speed obtained in the step (1) as two dimensions of clustering operation, and then carrying out clustering operation. And (3) enabling the number of the clustering centers to be iterated within the range of [2, 2ln n ], wherein n is the sample data size. Each clustering was normalized to the result of the objective function:
Figure BDA0001680752890000062
in the formula, U is a membership matrix, C is a cluster center set, m is a fuzzy weighting coefficient, n represents the data volume of sample data, C is the number of cluster centers, UijMembership, u, for the jth sample belonging to the ith cluster centerisMembership, x, for the ith sample belonging to the ith cluster centerjRepresents the jth sample data, ciRepresents the ith cluster center, | xj-ciAnd | | represents the Euclidean distance from the jth sample point to the ith cluster center.
Is defined as 10-6And when the value of the membership degree matrix is smaller than the threshold value, ending the clustering process.
The clustering process is a process of continuously iterative calculation of a membership matrix and a clustering center, and the membership matrix calculation formula is as follows:
Figure BDA0001680752890000071
in the formula, m is a fuzzy weighting coefficient, c is the number of clustering centers, ciDenotes the ith cluster center, ckDenotes the k-th cluster center, xjDenotes the jth sample data, uijMembership, u, for the jth sample belonging to the ith cluster centerisMembership, u, of the s-th sample to the i-th cluster centerrsMembership, u, of the s-th sample to the r-th cluster centerrjThe degree of membership of the jth sample belonging to the r-th cluster center.
The calculation formula of the clustering center is as follows:
Figure BDA0001680752890000072
in the formula, m is a fuzzy weighting coefficient, n represents the data volume of sample data, uijMembership, u, for the jth sample belonging to the ith cluster centerikMembership, x, for the kth sample belonging to the ith cluster centerjThe jth sample data is represented.
The specific clustering operation process is as follows:
i) let the number of cluster centers c be 2.
ii) initializing a membership matrix uijAt initialization time uijIs a value of [0,1]Random number within the range.
iii) calculating each cluster center c according to the membership matrixiThe value in each dimension.
iv) according to the clustering center ciCalculating a membership matrix uij。
v) if the objective function value is larger than the target function value, returning to the step iii, otherwise, performing the step vi.
vi) at this time, the clustering operation under the clustering center number c is completed, let c be c +1, if c is less than or equal to 2ln n, the adjacency matrix expression form of the clustering result under the current clustering center number is calculated, and the adjacency matrix meaning is as follows: if the element in the ith row and the jth column is 1, the ith sample data and the jth sample data belong to the same category, and the step ii is returned after the calculation of the adjacency matrix is completed. If c >2ln, proceed to step vii.
vii) at this time, the clustering operation under all the clustering center numbers is completed, the adjacent matrixes in the step vi of the clustering operation under each clustering center number are counted, all the adjacent matrixes are added to obtain a total adjacent matrix OJ. And counting the number of the clustering centers and the corresponding iteration times by using TreeMap, wherein the key of the TreeMap is the number of the clustering centers, and the value is the iteration times. Mixing O withJ Subtracting 1 from the value of not 0, and counting the current OJThe number num of the middle connected subgraphs is judged, whether a key with num exists in the TreeMap is judged, if not, the num is added into the TreeMap, and the corresponding value is 1; if a num key already exists in TreeMap, the value of the key is increased by 1 until OJAll elements in (1) are 0.
viii) selecting the key with the largest TreeMap median as the clustering center number, and taking the clustering result of the clustering center number as the clustering result of the sample data. If keys with the same value exist, judging by using a clustering effect evaluation index, respectively calculating the clustering effect evaluation index value under the condition that the keys with the same value are taken as the number of clustering centers, selecting the number of clustering centers with a smaller value of the clustering effect evaluation index as the number of selected clustering centers, wherein the calculation formula of the clustering effect evaluation index is as follows:
Figure BDA0001680752890000081
in the formula, c is the number of clustering centers, m is a fuzzy weighting coefficient, n represents the data volume of sample data, and uijMembership of the jth sample to the ith cluster center, ciDenotes the ith cluster center, ckDenotes the k-th cluster center, xjThe jth sample data is represented.
3) Dividing the total traffic jam state TC into 6 types of very smooth, unblocked, slow-moving, light jam, jam and serious jam, which respectively correspond to the total traffic jam state TC<TC1,TC2,…,TC6>The traffic congestion state occurring on all roads should be a subset of the TCs; taking the average speed as a vertical coordinate, taking the equivalent number of standard vehicles as a horizontal coordinate, sequencing the clustering centers from large to small according to the values of vertical coordinate projection points, projecting the clustering center points with the same values of the vertical coordinate projection points to the horizontal coordinate, sequencing the clustering center points from small to large according to the values of the horizontal coordinate projection points, representing the continuous change trend of the traffic jam state from unblocked to jammed, and then matching the traffic jam state of the clustering centers according to the leftmost state continuous matching principle, specifically comprising the following steps:
31) drawing the clustering center points on coordinate axes which respectively take the equivalent number of the standard vehicle and the average speed as horizontal and vertical coordinates;
32) sorting the clustering centers from large to small according to the values of the vertical coordinate projection points;
33) if the same vertical coordinate projection value exists (as shown in fig. 3), for the cluster center points with the same vertical coordinate projection value, the cluster center points are projected to the horizontal coordinate and are sorted from small to large according to the horizontal coordinate projection value, and the cluster center points corresponding to the sorted projection points represent the continuous change trend of the traffic jam state from smooth to jam.
34) Identifying the traffic jam state of the clustering center according to the principle of 'leftmost state continuous matching';
the principle of "leftmost state continuous matching": the sorted c clustering centers are always matched with the first c traffic jam states TC one by one from left to right.
35) And obtaining the traffic jam states of all sample points around all the clustering centers according to the traffic jam states of the clustering centers.
FIG. 4 shows an example of a continuous process of change of traffic congestion status, wherein TC1And TC6Respectively the lower bound and the upper bound of the total traffic jam state, and the traffic jam state at any time point is made to be TCiThen, the traffic jam state changes in the direction of increasing i (jam direction) or decreasing i (clear direction), and a continuous characteristic is presented, and the situation that the change of multi-level jumps does not occur is avoided, namely, each traffic jam state can only change towards the traffic jam state adjacent to the traffic jam state.
After the sorted clustering centers are obtained, according to the principle of 'leftmost state continuous matching', the traffic jam states represented by the clustering centers and the traffic jam state labels TC correspond one by one from left to right. Therefore, the cluster center and the traffic jam states of all the sample points belonging to the cluster center can be obtained.
Finally, the above embodiments are only for illustrating the technical solutions of the present invention and not for limiting, although the present invention has been described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that modifications or equivalent substitutions may be made on the technical solutions of the present invention without departing from the spirit and scope of the technical solutions, and all of them should be covered in the protection scope of the present invention.

Claims (3)

1. A method for judging the traffic jam state of an urban road based on RFID electronic license plate data is characterized by comprising the following steps:
1) acquiring traffic flow evaluation parameters including equivalent number and average speed of standard vehicles through RFID electronic license plate data;
2) taking the equivalent number and the average speed of the standard vehicles obtained in the step 1) as two dimensions of clustering operation, and performing clustering operation to obtain the optimal clustering number and the optimal clustering result, wherein the method specifically comprises the following steps:
21) let the number c of cluster centers be 2;
22) initializing membership matrix uijAt initialization time uijIs a value of [0,1]A random number within a range;
23) calculating each clustering center c according to the membership matrixiThe values in each dimension:
Figure FDA0002785298920000011
in the formula, m is a fuzzy weighting coefficient, n represents the data size of sample data, uijMembership, u, for the jth sample belonging to the ith cluster centerikMembership, x, for the kth sample belonging to the ith cluster centerjRepresenting jth sample data;
24) according to the cluster center ciCalculating a membership matrix uij：
Figure FDA0002785298920000012
In the formula, m is a fuzzy weighting coefficient, c is the number of clustering centers, ciDenotes the ith cluster center, ckDenotes the k-th cluster center, xjDenotes the jth sample data, uijMembership, u, for the jth sample belonging to the ith cluster centerisMembership, u, of the s-th sample to the i-th cluster centerrsMembership, u, of the s-th sample to the r-th cluster centerrjThe membership degree of the jth sample belonging to the r clustering center;
25) if the objective function value is larger than the target function value, returning to the step 23), otherwise, performing the step 26);
26) at this time, the clustering operation under the clustering center number c is completed, let c be c +1, if c be less than or equal to 2ln n, the adjacency matrix expression form of the clustering result under the current clustering center number is calculated, and the adjacency matrix meaning is as follows: if the element of the ith row and the jth column is 1, the ith sample data and the jth sample data belong to the same category, and the step 22) is returned after the calculation of the adjacency matrix is completed; if c is more than 2ln n, performing the step 27);
27) at this time, the clustering operation is completed under all the clustering center numbers, the adjacent matrixes in the 26) step of the clustering operation under each clustering center number are counted, all the adjacent matrixes are added to obtain a total adjacent matrix OJ(ii) a Utilizing TreeMap to count the number of the clustering centers and the corresponding iteration times, and enabling the key of TreeMap to be the number of the clustering centers and the value to be the iteration times; mixing O withJSubtracting 1 from the value of not 0, and counting the current OJThe number num of the middle connected subgraphs is judged, whether a key with num exists in the TreeMap is judged, if not, the num is added into the TreeMap, and the corresponding value is 1; if a num key already exists in TreeMap, the value of the key is increased by 1 until OJAll elements in the formula are 0;
28) selecting a key with the largest TreeMap median as the number of clustering centers, and taking the clustering result of the number of the clustering centers as the clustering result of the sample data; if keys with the same value exist, judging by using a clustering effect evaluation index, respectively calculating the clustering effect evaluation index value under the condition that the keys with the same value are taken as the number of clustering centers, selecting the number of clustering centers with a smaller value of the clustering effect evaluation index as the number of selected clustering centers, wherein the calculation formula of the clustering effect evaluation index is as follows:
Figure FDA0002785298920000021
in the formula, c is the number of clustering centers, m is a fuzzy weighting coefficient, n represents the data volume of sample data, and uijMembership of the jth sample to the ith cluster center, ciDenotes the ith cluster center, ckDenotes the k-th cluster center, xjRepresenting jth sample data;
3) taking the average speed as a vertical coordinate, taking the equivalent number of standard vehicles as a horizontal coordinate, sequencing the clustering centers from large to small according to the values of vertical coordinate projection points, projecting the clustering center points with the same values of the vertical coordinate projection points to the horizontal coordinate, sequencing according to the values of the horizontal coordinate projection points from small to large, representing the continuous change trend of the traffic jam state from unblocked to jammed by the clustering center points corresponding to the sequenced projection points, and then matching the traffic jam state of the clustering centers according to the leftmost state continuous matching principle.
2. The method for distinguishing the traffic jam state of the urban road based on the RFID electronic license plate data as claimed in claim 1, wherein the step 1) specifically comprises the following steps:
11) the RFID acquisition points acquire RFID electronic license plate data to obtain the traffic volume of all vehicles and the traffic volume of large vehicles in a time period, and the equivalent number of standard vehicles in the time period is calculated according to the following formula:
pcu＝n+n2；
wherein pcu represents the number of equivalents of standard vehicle, n represents the traffic volume of all vehicle types, n2Representing the traffic volume of a large vehicle;
12) the average speed of the vehicle is calculated by:
Figure FDA0002785298920000031
in the formula, v represents the average speed in km/h, L represents the distance between two acquisition points of the road section in meters, ti1And ti2Respectively representing the time of the ith vehicle passing through the first RFID acquisition point and the second RFID acquisition point, wherein the unit is second, and n represents the traffic volume of all vehicle types in the time period.
3. The method for distinguishing the traffic jam state of the urban road based on the RFID electronic license plate data according to claim 1 or 2, wherein the step 3) specifically comprises the following steps:
31) drawing the clustering center points on coordinate axes which respectively take the equivalent number of the standard vehicle and the average speed as horizontal and vertical coordinates;
32) sorting the clustering centers from large to small according to the values of the vertical coordinate projection points;
33) if the same vertical coordinate projection values exist, for the clustering center points with the same vertical coordinate projection values, the clustering center points are projected to the horizontal coordinate and are sorted from small to large according to the horizontal coordinate projection values, and the clustering center points corresponding to the sorted projection points represent the continuous change trend of the traffic jam state from unblocked to jammed;
34) identifying the traffic jam state of the clustering center according to a leftmost state continuous matching principle; the leftmost state continuous matching principle is as follows: c sequenced clustering centers are always matched with the front c traffic jam states TC one by one from left to right; the traffic jam state TC is divided into 6 types of very smooth, unblocked, slow-moving, light jam, jam and serious jam which respectively correspond to the traffic jam state TC<TC1,TC2,…,TC6>；
35) And obtaining the traffic jam states of all sample points around all the clustering centers according to the traffic jam states of the clustering centers.
CN201810549401.0A 2018-05-31 2018-05-31 Urban road traffic jam state discrimination method based on RFID electronic license plate data Active CN108682153B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201810549401.0A CN108682153B (en) 2018-05-31 2018-05-31 Urban road traffic jam state discrimination method based on RFID electronic license plate data

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201810549401.0A CN108682153B (en) 2018-05-31 2018-05-31 Urban road traffic jam state discrimination method based on RFID electronic license plate data

Publications (2)

Publication Number Publication Date
CN108682153A CN108682153A (en) 2018-10-19
CN108682153B true CN108682153B (en) 2020-12-29

Family

ID=63809412

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201810549401.0A Active CN108682153B (en) 2018-05-31 2018-05-31 Urban road traffic jam state discrimination method based on RFID electronic license plate data

Country Status (1)

Country Link
CN (1) CN108682153B (en)

Families Citing this family (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109785622B (en) * 2018-12-03 2020-12-11 南京三宝科技股份有限公司 Congestion area identification method based on vehicle-mounted electronic tag technology
CN111613049B (en) * 2019-02-26 2022-07-12 北京嘀嘀无限科技发展有限公司 Road state monitoring method and device
CN112330953A (en) * 2020-10-10 2021-02-05 杭州翔毅科技有限公司 Traffic control method, device and equipment based on cloud data and storage medium
CN112767684A (en) * 2020-12-22 2021-05-07 重庆大学 Highway traffic jam detection method based on charging data

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103500504A (en) * 2013-09-30 2014-01-08 同济大学 Urban road traffic parameter estimation and road situation discrimination method based on RFID (Radio Frequency Identification) data and application system
CN104517113A (en) * 2013-09-29 2015-04-15 浙江大华技术股份有限公司 Image feature extraction method and device and image sorting method and device
CN106056923A (en) * 2016-06-27 2016-10-26 安徽四创电子股份有限公司 Road traffic state discrimination method based on traffic scene radar
CN106530709A (en) * 2016-12-16 2017-03-22 东南大学 User-oriented highway traffic index publishing system
CN107153846A (en) * 2017-05-26 2017-09-12 南京邮电大学 A kind of road traffic state modeling method based on Fuzzy C-Means Cluster Algorithm
WO2017174623A1 (en) * 2016-04-05 2017-10-12 Tomtom Traffic B.V. Method and apparatus for identifying congestion bottlenecks
CN107705551A (en) * 2016-08-08 2018-02-16 阿里巴巴集团控股有限公司 The appraisal procedure and device of traffic

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102546771A (en) * 2011-12-27 2012-07-04 西安博构电子信息科技有限公司 Cloud mining network public opinion monitoring system based on characteristic model

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104517113A (en) * 2013-09-29 2015-04-15 浙江大华技术股份有限公司 Image feature extraction method and device and image sorting method and device
CN103500504A (en) * 2013-09-30 2014-01-08 同济大学 Urban road traffic parameter estimation and road situation discrimination method based on RFID (Radio Frequency Identification) data and application system
WO2017174623A1 (en) * 2016-04-05 2017-10-12 Tomtom Traffic B.V. Method and apparatus for identifying congestion bottlenecks
CN106056923A (en) * 2016-06-27 2016-10-26 安徽四创电子股份有限公司 Road traffic state discrimination method based on traffic scene radar
CN107705551A (en) * 2016-08-08 2018-02-16 阿里巴巴集团控股有限公司 The appraisal procedure and device of traffic
CN106530709A (en) * 2016-12-16 2017-03-22 东南大学 User-oriented highway traffic index publishing system
CN107153846A (en) * 2017-05-26 2017-09-12 南京邮电大学 A kind of road traffic state modeling method based on Fuzzy C-Means Cluster Algorithm

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
基于FCM聚类的多路段城市道路车流数据分析及其应用;王旭琴 等;《系统工程理论与应用》;20141231;第3-13页 *

Also Published As

Publication number Publication date
CN108682153A (en) 2018-10-19

Similar Documents

Publication Publication Date Title
CN108682153B (en) Urban road traffic jam state discrimination method based on RFID electronic license plate data
CN101504717B (en) Characteristic area positioning method, car body color depth and color recognition method
CN110728842B (en) Abnormal driving early warning method based on reasonable driving range of vehicles at intersection
CN101216998B (en) An urban traffic flow information amalgamation method of evidence theory based on fuzzy rough sets
CN105335702B (en) A kind of bayonet model recognizing method based on statistical learning
CN101329734B (en) License plate character recognition method based on K-L transform and LS-SVM
CN103366367B (en) Based on the FCM gray-scale image segmentation method of pixel count cluster
CN112016605B (en) Target detection method based on corner alignment and boundary matching of bounding box
CN110458244A (en) A kind of traffic accident Severity forecasting method applied to Regional Road Network
CN107301433A (en) Net based on clustering and discriminant model about car discrimination method and system
CN105809193B (en) A kind of recognition methods of the illegal vehicle in use based on kmeans algorithm
CN108830254B (en) Fine-grained vehicle type detection and identification method based on data balance strategy and intensive attention network
CN107845260B (en) Method for identifying public transport trip mode of user
CN104951764A (en) Identification method for behaviors of high-speed vehicle based on secondary spectrum clustering and HMM (Hidden Markov Model)-RF (Random Forest) hybrid model
CN104182985A (en) Remote sensing image change detection method
CN103679214B (en) Vehicle checking method based on online Class area estimation and multiple features Decision fusion
CN108447265A (en) Road traffic accident stain section discrimination method based on TOPSIS methods
CN102184412B (en) Licence plate number and letter identification method based on minimum-error-rate Bayes classifier
CN110705484A (en) Method for recognizing illegal behavior of continuously changing lane by using driving track
CN112767684A (en) Highway traffic jam detection method based on charging data
Liu et al. A large-scale benchmark for vehicle logo recognition
CN105404858A (en) Vehicle type recognition method based on deep Fisher network
CN110991442A (en) High-accuracy highway license plate cloud identification method
CN101702172A (en) Data discretization method based on category-attribute relation dependency
CN106529391A (en) Robust speed-limit traffic sign detection and recognition method

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant