CN104361349A - Car inspection device and toll data fusion based abnormal traffic state identification method and system - Google Patents
Car inspection device and toll data fusion based abnormal traffic state identification method and system Download PDFInfo
- Publication number
- CN104361349A CN104361349A CN201410604146.7A CN201410604146A CN104361349A CN 104361349 A CN104361349 A CN 104361349A CN 201410604146 A CN201410604146 A CN 201410604146A CN 104361349 A CN104361349 A CN 104361349A
- Authority
- CN
- China
- Prior art keywords
- algorithm
- data
- trust
- recognition
- level
- 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.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Computational Biology (AREA)
- General Engineering & Computer Science (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Traffic Control Systems (AREA)
- Devices For Checking Fares Or Tickets At Control Points (AREA)
Abstract
The invention discloses a car inspection device and toll data fusion based abnormal traffic state identification method and system. The method includes: firstly, collecting road data through a car inspection device and identifying abnormal state data of the car inspection device by adopting an ACI (automatic car identification) method; secondly, acquiring abnormal road toll state data through road toll data; finally, identifying the abnormal road toll state data by adopting an algorithm voting fusion method. With the method, the abnormal traffic state identification problem that a single data source algorithm is low in credibility and poor in actual application effect is solved, requirements of highway traffic flow on complexity and randomicity on time and space are met, multiple information sources are comprehensively utilized, the abnormal traffic state is identified form multiple aspects, and the abnormal traffic state is more accurately identified.
Description
Technical field
The present invention relates to traffic status of express way identification field, be specifically related to a kind of abnormality recognition methods of merging based on vehicle checker and charge data.
Background technology
Abnormality identification is that highway Yun Guan department carries out runing management and control, traffic information is issued and the basis of traffic guidance, to reducing personal injury, property loss that traffic hazard causes and avoiding the aspects such as secondary traffic accident to have important effect.Along with the development of intelligent transport technology and the increase of transport information detection means, how to utilize transport information detection means to carry out abnormality and automatically identify and attract wide attention.Therefore, the gordian technique that research highway abnormality identifies automatically has important theoretical and practical significance to Improving Expressway management level and service level.
The identification of existing highway abnormality mainly, based on data mapping (vehicle checker data), is set up abnormality automatic identification algorithm, is automatically identified abnormal traffic state.Due to freeway traffic flow complicacy in time and spatially and randomness, under many circumstances, the problems such as with a low credibility, practical application effect is poor may be there is in ACI (the automatically identifying) algorithm based on data mapping, if simply algorithm bad for some differentiation effects is differentiated that result is directly cast out, this is obviously irrational.By contrast, a kind of way of more science is, is merged in some way by the ACI algorithm based on different pieces of information, comprehensive utilization Multiple Information Sources, from many aspects, abnormal traffic state is identified, so this patent adopts vehicle checker and charge data fusion recognition abnormal traffic state.
Summary of the invention
In view of this, technical matters to be solved by this invention is to provide a kind of abnormality recognition methods of merging based on vehicle checker and charge data and system.
An object of the present invention proposes a kind of abnormality recognition methods of merging based on vehicle checker and charge data; Two of object of the present invention proposes a kind of abnormality recognition system merged based on vehicle checker and charge data.
An object of the present invention is achieved through the following technical solutions:
A kind of abnormal traffic state recognition methods of merging based on vehicle checker and charge data provided by the invention, comprises the following steps:
S1: gather highway data extraction rate, flow and occupation rate traffic behavior characteristic parameter by vehicle checker;
S2: adopt ACI algorithm identification highway vehicle checker abnormality data according to traffic behavior characteristic parameter;
S3: obtain highway toll data and respective stretch traffic behavior parameter;
S4: calculate road-section average travel speed according to highway toll data and respective stretch traffic behavior parameter;
S5: obtain highway toll abnormality data with comparing of the predetermined average stroke speed of a motor vehicle according to road-section average travel speed;
S6: adopt algorithm voting fusion method according to highway vehicle checker abnormality data and highway toll abnormality data identification highway abnormality data.
Further, the traffic behavior characteristic parameter in described step S1 adopts California algorithm, and concrete implementation step is as follows:
S11: obtain upstream and downstream wagon detector data, calculate occupation rate difference (absolute difference):
OCCD=OCC(i,t)-OCC(i+1,t) (1)
In formula, OCCD represents the occupation rate absolute difference of adjacent two vehicle checkers; OCC (i, t) represents upstream vehicle checker occupation rate; OCC (i+1, t) represents downstream vehicle checker occupation rate;
S12: as upstream and downstream difference OCCD < K
1time, can be judged to be now non-blocking up, be normal traffic states; If OCCD>=K
1time, continue next step S13;
Wherein, K
1for the parameter value demarcated by experience;
S13: according to the occupation rate difference between upstream and downstream website, calculates its ratio with upstream stations occupation rate:
S13: as OCCDF < K
2time, can be judged to be now non-blocking up, be normal traffic states; If OCCDF>=K
2time, continue Step4; Wherein, K
2for the parameter value demarcated by experience;
S14: according to the occupation rate difference between upstream and downstream website, calculates its ratio with downstream website occupation rate:
S15: as DOCCDT > K
3time, can be judged to be now non-blocking up, be normal traffic states; If DOCCDT≤K
3time, continue S16; Wherein, K
3for the parameter value demarcated by experience;
S16: read a upper recognition cycle abnormal traffic state, if blocked up, then differentiate that this recognition cycle is that traffic behavior is abnormal, otherwise be normal traffic states, this recognition cycle terminates.
Further, the calculating of the highway toll abnormality data in described step S5, concrete implementation step is as follows:
S51: calculate the average stroke speed of a motor vehicle according to charge data
S52: according to different sections of highway feature, determines discriminant parameter K
1', K
2', K
3' value;
S53: determine traffic behavior according to the average stroke speed of a motor vehicle and discriminant parameter, determines according to under type:
S54: judge
whether set up, if set up, then show that this road section traffic volume is unimpeded; If be false, enter discriminating step S55;
S55: judge
whether set up, if set up, then show to walk or drive slowly in this section, namely travel speed bends down than unimpeded state, but than high under Traffic Congestion; If be false, enter discriminating step S56;
S56: judge
whether set up, if set up, then show that crowded character is now incidental congestion, if be false, then show to be now recurrent congestion.
Further, the algorithm voting fusion method in described step S6, concrete implementation step is as follows:
S61: determine that algorithm forms, described algorithm composition comprises the first algorithm and the second algorithm, and described first algorithm is the ACI algorithm based on vehicle checker data; Described second algorithm is the ACI algorithm based on charge data;
S62: the level of trust determining algorithm, continuation test is adopted to define algorithm level of trust, described continuation test is divided into Three Estate: PT=1,2,3, often kind of algorithm all adopts 3 grades of reliability ratings, in continuance test, as PT=1, represent that ACI algorithm provides abnormal traffic state recognition result through a recognition cycle; As PT=2, represent that ACI algorithm provides abnormal traffic state recognition result through two recognition cycles; As PT=3, represent that the duration of continuous 3 or more identifies abnormal traffic state, PT=3 is highest level of trust;
S63: determine recognition unit, described recognition unit is defined as follows:
Determine that the level of trust of the first algorithm is respectively the first level of trust A
1, the second level of trust A
2, the 3rd level of trust A
3;
Determine that the level of trust of the second algorithm is respectively the first level of trust B
1, the second level of trust B
2, the 3rd level of trust B
3;
Determine the first recognition unit A
1b
1, the second recognition unit A
2b
2, the 3rd recognition unit A
3b
3;
Described first recognition unit A
1b
1represent that the first algorithm is the first level of trust A
1be the first level of trust B with the second algorithm
1time the recognition unit determined;
Described second recognition unit A
2b
2represent that the first algorithm is the second level of trust A
2be the second level of trust B with the second algorithm
2time the recognition unit determined;
Described 3rd recognition unit A
3b
3represent that the first algorithm is the 3rd level of trust A
3be the 3rd level of trust B with the second algorithm
3time the recognition unit determined;
S64: the recognition time coupling determining the first algorithm and the second algorithm recognition result, certainty annuity clock synchronous, described system clock synchronization comprise the ACI algoritic module based on charge data system clock, based on the system clock of ACI algoritic module of vehicle checker data and the system clock of Fusion Module; Concrete matching principle is as follows:
The data collection cycle of setting vehicle checker is identical with charge data collection period;
When the ACI algorithm based on charge data is consistent with the ACI algorithm recognition time based on vehicle checker data, the ACI algorithm recognition time set based on charge data is identical with the ACI algorithm recognition time based on vehicle checker data;
When the ACI algorithm based on charge data and the ACI algorithm recognition time based on vehicle checker data inconsistent time, postpone to identify the recognition time of preceding a kind of algorithm.
Two of object of the present invention is achieved through the following technical solutions:
A kind of abnormal traffic state recognition system merged based on vehicle checker and charge data provided by the invention, comprises the ACI algoritic module based on charge data, the ACI algoritic module based on vehicle checker data and Fusion Module;
The described ACI algoritic module based on charge data is used for calculating average travel speed according to charge data traffic behavior parameter and compared with the predetermined average stroke speed of a motor vehicle, thus judges highway toll abnormality data;
The described ACI algoritic module based on vehicle checker data is used for determining highway vehicle checker abnormality data according to the traffic behavior parameter of vehicle checker data acquisition;
Described Fusion Module is used for the highway toll abnormality data obtained by the ACI algoritic module based on charge data and the abnormal traffic state identification signal calculating highway based on the highway vehicle checker abnormality data acquisition algorithm voting fusion method that the ACI algoritic module of vehicle checker data obtains.
Further, described Fusion Module comprises and determines algorithm component units, determines algorithm level of trust unit, determines recognition unit and recognition time matching unit;
Describedly determine that algorithm component units is for determining the algorithm ingredient needing to merge, described algorithm composition comprises the first algorithm and the second algorithm, and described first algorithm is the ACI algorithm based on vehicle checker data; Described second algorithm is the ACI algorithm based on charge data;
Describedly determine that algorithm level of trust unit is for determining the level of trust of algorithm ingredient, described algorithm level of trust adopts continuation test to define algorithm level of trust, described continuation test is divided into Three Estate: PT=1,2,3, often kind of algorithm all adopts 3 grades of reliability ratings, in continuance test, as PT=1, represent that ACI algorithm provides abnormal traffic state recognition result through a recognition cycle; As PT=2, represent that ACI algorithm provides abnormal traffic state recognition result through two recognition cycles; As PT=3, represent that the duration of continuous 3 or more identifies abnormal traffic state, PT=3 is highest level of trust;
Describedly determine that recognition unit is for determining according to algorithm level of trust and algorithm ingredient the recognition result that algorithm forms, described recognition unit is defined as follows:
Determine that the level of trust of the first algorithm is respectively the first level of trust A
1, the second level of trust A
2, the 3rd level of trust A
3;
Determine that the level of trust of the second algorithm is respectively the first level of trust B
1, the second level of trust B
2, the 3rd level of trust B
3;
Determine the first recognition unit A
1b
1, the second recognition unit A
2b
2, the 3rd recognition unit A
3b
3;
Described first recognition unit A
1b
1represent that the first algorithm is the first level of trust A
1be the first level of trust B with the second algorithm
1time the recognition unit determined;
Described second recognition unit A
2b
2represent that the first algorithm is the second level of trust A
2be the second level of trust B with the second algorithm
2time the recognition unit determined;
Described 3rd recognition unit A
3b
3represent that the first algorithm is the 3rd level of trust A
3be the 3rd level of trust B with the second algorithm
3time the recognition unit determined;
Described recognition time matching unit is used for the recognition time of unified algorithms of different composition, for determining the recognition time coupling of the first algorithm and the second algorithm recognition result, certainty annuity clock synchronous, described system clock synchronization comprise the ACI algoritic module based on charge data system clock, based on the system clock of ACI algoritic module of vehicle checker data and the system clock of Fusion Module; Concrete matching principle is as follows:
The data collection cycle of setting vehicle checker is identical with charge data collection period;
When the ACI algorithm based on charge data is consistent with the ACI algorithm recognition time based on vehicle checker data, the ACI algorithm recognition time set based on charge data is identical with the ACI algorithm recognition time based on vehicle checker data;
When the ACI algorithm based on charge data and the ACI algorithm recognition time based on vehicle checker data inconsistent time, postpone to identify the recognition time of preceding a kind of algorithm.
Further, described acquisition based on the traffic behavior characteristic parameter in the ACI algoritic module of vehicle checker data adopts California algorithm, and concrete implementation step is as follows:
S11: obtain upstream and downstream wagon detector data, calculate occupation rate difference (absolute difference):
OCCD=OCC(i,t)-OCC(i+1,t) (1)
In formula, OCCD represents the occupation rate absolute difference of adjacent two vehicle checkers; OCC (i, t) represents upstream vehicle checker occupation rate; OCC (i+1, t) represents downstream vehicle checker occupation rate;
S12: as upstream and downstream difference OCCD < K
1time, can be judged to be now non-blocking up, be normal traffic states; If OCCD>=K
1time, continue next step S13;
Wherein, K
1, K
2, K
3for the parameter value demarcated by experience;
S13: according to the occupation rate difference between upstream and downstream website, calculates its ratio with upstream stations occupation rate:
S13: as OCCDF < K
2time, can be judged to be now non-blocking up, be normal traffic states; If OCCDF>=K
2time, continue Step4; Wherein, K
1, K
2, K
3for the parameter value demarcated by experience;
S14: according to the occupation rate difference between upstream and downstream website, calculates its ratio with downstream website occupation rate:
S15: as DOCCDT > K
3time, can be judged to be now non-blocking up, be normal traffic states; If DOCCDT≤K
3time, continue S16; Wherein, K
3for the parameter value demarcated by experience;
S16: read a upper recognition cycle abnormal traffic state, if blocked up, then differentiate that this recognition cycle is that traffic behavior is abnormal, otherwise be normal traffic states, this recognition cycle terminates.
Further, the described calculating based on the highway toll abnormality data in the ACI algoritic module of charge data, concrete implementation step is as follows:
S51: calculate the average stroke speed of a motor vehicle according to charge data
S52: according to different sections of highway feature, determines discriminant parameter K
1', K
2', K
3' value;
S53: determine traffic behavior according to the average stroke speed of a motor vehicle and discriminant parameter, determines according to under type:
S54: judge
whether set up, if set up, then show that this road section traffic volume is unimpeded; If be false, enter discriminating step S55;
S55: judge
whether set up, if set up, then show to walk or drive slowly in this section, namely travel speed bends down than unimpeded state, but than high under Traffic Congestion; If be false, enter discriminating step S56;
S56: judge
whether set up, if set up, then show that crowded character is now incidental congestion, if be false, then show to be now recurrent congestion.
Beneficial effect of the present invention is: the present invention adopts the abnormal traffic state recognition methods of merging based on vehicle checker and charge data, overcome the abnormal traffic state identification problem that data mapping algorithm is with a low credibility, practical application effect is poor, the present invention adopts charge data and vehicle checker data to set up abnormality automatic identification algorithm respectively, then adopts the foundation of algorithm voting fusion method based on the abnormal traffic state recognition methods in different pieces of information source.Be applicable to freeway traffic flow complicacy in time and spatially and the requirement of randomness, comprehensive utilization Multiple Information Sources, identifies from many aspects abnormal traffic state, identifies that abnormal traffic state is more accurate.
Accompanying drawing explanation
In order to make the object, technical solutions and advantages of the present invention clearly, below in conjunction with accompanying drawing, the present invention is described in further detail, wherein:
The highway abnormality identification block diagram merged based on vehicle checker and charge data that Fig. 1 provides for the embodiment of the present invention;
The California algorithm flow chart that Fig. 2 provides for the embodiment of the present invention;
The ACI algorithm flow chart based on charge data that Fig. 3 provides for the embodiment of the present invention;
The hardware implementing of the decision-making voting fusion method system logic that Fig. 4 provides for the embodiment of the present invention.
Embodiment
Hereinafter with reference to accompanying drawing, the preferred embodiments of the present invention are described in detail.Should be appreciated that preferred embodiment only in order to the present invention is described, instead of in order to limit the scope of the invention.
Embodiment 1
As shown in Figures 1 to 4: a kind of abnormal traffic state recognition methods of merging based on vehicle checker and charge data provided by the invention, comprises the following steps:
S1: gather highway data extraction rate, flow and occupation rate traffic behavior characteristic parameter by vehicle checker;
S2: adopt ACI algorithm identification highway vehicle checker abnormality data according to traffic behavior characteristic parameter;
S3: obtain highway toll data and respective stretch traffic behavior parameter;
S4: calculate road-section average travel speed according to highway toll data and respective stretch traffic behavior parameter;
S5: obtain highway toll abnormality data with comparing of the predetermined average stroke speed of a motor vehicle according to road-section average travel speed;
S6: adopt algorithm voting fusion method according to highway vehicle checker abnormality data and highway toll abnormality data identification highway abnormality data.
Traffic behavior characteristic parameter in described step S1 adopts California algorithm, and concrete implementation step is as follows:
S11: obtain upstream and downstream wagon detector data, calculate occupation rate difference (absolute difference):
OCCD=OCC(i,t)-OCC(i+1,t) (1)
In formula, OCCD represents the occupation rate absolute difference of adjacent two vehicle checkers; OCC (i, t) represents upstream vehicle checker occupation rate; OCC (i+1, t) represents downstream vehicle checker occupation rate;
S12: as upstream and downstream difference OCCD < K
1time, can be judged to be now non-blocking up, be normal traffic states; If OCCD>=K
1time, continue next step S13;
Wherein, K
1for the parameter value demarcated by experience;
S13: according to the occupation rate difference between upstream and downstream website, calculates its ratio with upstream stations occupation rate:
S13: as OCCDF < K
2time, can be judged to be now non-blocking up, be normal traffic states; If OCCDF>=K
2time, continue Step4; Wherein, K
2for the parameter value demarcated by experience;
S14: according to the occupation rate difference between upstream and downstream website, calculates its ratio with downstream website occupation rate:
S15: as DOCCDT > K
3time, can be judged to be now non-blocking up, be normal traffic states; If DOCCDT≤K
3time, continue S16; Wherein, K
3for the parameter value demarcated by experience;
S16: read a upper recognition cycle abnormal traffic state, if blocked up, then differentiate that this recognition cycle is that traffic behavior is abnormal, otherwise be normal traffic states, this recognition cycle terminates.
The calculating of the highway toll abnormality data in described step S5, concrete implementation step is as follows:
S51: calculate the average stroke speed of a motor vehicle according to charge data
S52: according to different sections of highway feature, determines discriminant parameter K
1', K
2', K
3' value;
S53: determine traffic behavior according to the average stroke speed of a motor vehicle and discriminant parameter, determines according to under type:
S54: judge
whether set up, if set up, then show that this road section traffic volume is unimpeded; If be false, enter discriminating step S55;
S55: judge
whether set up, if set up, then show to walk or drive slowly in this section, namely travel speed bends down than unimpeded state, but than high under Traffic Congestion; If be false, enter discriminating step S56;
S56: judge
whether set up, if set up, then show that crowded character is now incidental congestion, if be false, then show to be now recurrent congestion.
Algorithm voting fusion method in described step S6, concrete implementation step is as follows:
S61: determine that algorithm forms, described algorithm composition comprises the first algorithm and the second algorithm, and described first algorithm is the ACI algorithm based on vehicle checker data; Described second algorithm is the ACI algorithm based on charge data;
S62: the level of trust determining algorithm, continuation test is adopted to define algorithm level of trust, described continuation test is divided into Three Estate: PT=1,2,3, often kind of algorithm all adopts 3 grades of reliability ratings, in continuance test, as PT=1, represent that ACI algorithm provides abnormal traffic state recognition result through a recognition cycle; As PT=2, represent that ACI algorithm provides abnormal traffic state recognition result through two recognition cycles; As PT=3, represent that the duration of continuous 3 or more identifies abnormal traffic state, PT=3 is highest level of trust;
S63: determine recognition unit, described recognition unit is defined as follows:
Determine that the level of trust of the first algorithm is respectively the first level of trust A
1, the second level of trust A
2, the 3rd level of trust A
3;
Determine that the level of trust of the second algorithm is respectively the first level of trust B
1, the second level of trust B
2, the 3rd level of trust B
3;
Determine the first recognition unit A
1b
1, the second recognition unit A
2b
2, the 3rd recognition unit A
3b
3;
Described first recognition unit A
1b
1represent that the first algorithm is the first level of trust A
1be the first level of trust B with the second algorithm
1time the recognition unit determined;
Described second recognition unit A
2b
2represent that the first algorithm is the second level of trust A
2be the second level of trust B with the second algorithm
2time the recognition unit determined;
Described 3rd recognition unit A
3b
3represent that the first algorithm is the 3rd level of trust A
3be the 3rd level of trust B with the second algorithm
3time the recognition unit determined;
S64: the recognition time coupling determining the first algorithm and the second algorithm recognition result, certainty annuity clock synchronous, described system clock synchronization comprise the ACI algoritic module based on charge data system clock, based on the system clock of ACI algoritic module of vehicle checker data and the system clock of Fusion Module; Concrete matching principle is as follows:
The data collection cycle of setting vehicle checker is identical with charge data collection period;
When the ACI algorithm based on charge data is consistent with the ACI algorithm recognition time based on vehicle checker data, the ACI algorithm recognition time set based on charge data is identical with the ACI algorithm recognition time based on vehicle checker data;
When the ACI algorithm based on charge data and the ACI algorithm recognition time based on vehicle checker data inconsistent time, postpone to identify the recognition time of preceding a kind of algorithm.
The present embodiment additionally provides a kind of abnormal traffic state recognition system merged based on vehicle checker and charge data, comprises the ACI algoritic module based on charge data, the ACI algoritic module based on vehicle checker data and Fusion Module;
The described ACI algoritic module based on charge data is used for calculating average travel speed according to charge data traffic behavior parameter and compared with the predetermined average stroke speed of a motor vehicle, thus judges highway toll abnormality data;
The described ACI algoritic module based on vehicle checker data is used for determining highway vehicle checker abnormality data according to the traffic behavior parameter of vehicle checker data acquisition;
Described Fusion Module is used for the highway toll abnormality data obtained by the ACI algoritic module based on charge data and the abnormal traffic state identification signal calculating highway based on the highway vehicle checker abnormality data acquisition algorithm voting fusion method that the ACI algoritic module of vehicle checker data obtains.
Described Fusion Module comprises to be determined algorithm component units, determine algorithm level of trust unit, determines recognition unit and recognition time matching unit;
Describedly determine that algorithm component units is for determining the algorithm ingredient needing to merge, described algorithm composition comprises the first algorithm and the second algorithm, and described first algorithm is the ACI algorithm based on vehicle checker data; Described second algorithm is the ACI algorithm based on charge data;
Describedly determine that algorithm level of trust unit is for determining the level of trust of algorithm ingredient, described algorithm level of trust adopts continuation test to define algorithm level of trust, described continuation test is divided into Three Estate: PT=1,2,3, often kind of algorithm all adopts 3 grades of reliability ratings, in continuance test, as PT=1, represent that ACI algorithm provides abnormal traffic state recognition result through a recognition cycle; As PT=2, represent that ACI algorithm provides abnormal traffic state recognition result through two recognition cycles; As PT=3, represent that the duration of continuous 3 or more identifies abnormal traffic state, PT=3 is highest level of trust;
Describedly determine that recognition unit is for determining according to algorithm level of trust and algorithm ingredient the recognition result that algorithm forms, described recognition unit is defined as follows:
Determine that the level of trust of the first algorithm is respectively the first level of trust A
1, the second level of trust A
2, the 3rd level of trust A
3;
Determine that the level of trust of the second algorithm is respectively the first level of trust B
1, the second level of trust B
2, the 3rd level of trust B
3;
Determine the first recognition unit A
1b
1, the second recognition unit A
2b
2, the 3rd recognition unit A
3b
3;
Described first recognition unit A
1b
1represent that the first algorithm is the first level of trust A
1be the first level of trust B with the second algorithm
1time the recognition unit determined;
Described second recognition unit A
2b
2represent that the first algorithm is the second level of trust A
2be the second level of trust B with the second algorithm
2time the recognition unit determined;
Described 3rd recognition unit A
3b
3represent that the first algorithm is the 3rd level of trust A
3be the 3rd level of trust B with the second algorithm
3time the recognition unit determined;
Described recognition time matching unit is used for the recognition time of unified algorithms of different composition, for determining the recognition time coupling of the first algorithm and the second algorithm recognition result, certainty annuity clock synchronous, described system clock synchronization comprise the ACI algoritic module based on charge data system clock, based on the system clock of ACI algoritic module of vehicle checker data and the system clock of Fusion Module; Concrete matching principle is as follows:
The data collection cycle of setting vehicle checker is identical with charge data collection period;
When the ACI algorithm based on charge data is consistent with the ACI algorithm recognition time based on vehicle checker data, the ACI algorithm recognition time set based on charge data is identical with the ACI algorithm recognition time based on vehicle checker data;
When the ACI algorithm based on charge data and the ACI algorithm recognition time based on vehicle checker data inconsistent time, postpone to identify the recognition time of preceding a kind of algorithm.
Described acquisition based on the traffic behavior characteristic parameter in the ACI algoritic module of vehicle checker data adopts California algorithm, and concrete implementation step is as follows:
S11: obtain upstream and downstream wagon detector data, calculate occupation rate difference (absolute difference):
OCCD=OCC(i,t)-OCC(i+1,t) (1)
In formula, OCCD represents the occupation rate absolute difference of adjacent two vehicle checkers; OCC (i, t) represents upstream vehicle checker occupation rate; OCC (i+1, t) represents downstream vehicle checker occupation rate;
S12: as upstream and downstream difference OCCD < K
1time, can be judged to be now non-blocking up, be normal traffic states; If OCCD>=K
1time, continue next step S13;
Wherein, K
1, K
2, K
3for the parameter value demarcated by experience;
S13: according to the occupation rate difference between upstream and downstream website, calculates its ratio with upstream stations occupation rate:
S13: as OCCDF < K
2time, can be judged to be now non-blocking up, be normal traffic states; If OCCDF>=K
2time, continue Step4; Wherein, K
1, K
2, K
3for the parameter value demarcated by experience;
S14: according to the occupation rate difference between upstream and downstream website, calculates its ratio with downstream website occupation rate:
S15: as DOCCDT > K
3time, can be judged to be now non-blocking up, be normal traffic states; If DOCCDT≤K
3time, continue S16; Wherein, K
3for the parameter value demarcated by experience;
S16: read a upper recognition cycle abnormal traffic state, if blocked up, then differentiate that this recognition cycle is that traffic behavior is abnormal, otherwise be normal traffic states, this recognition cycle terminates.
The described calculating based on the highway toll abnormality data in the ACI algoritic module of charge data, concrete implementation step is as follows:
S51: calculate the average stroke speed of a motor vehicle according to charge data
S52: according to different sections of highway feature, determines discriminant parameter K
1', K
2', K
3' value;
S53: determine traffic behavior according to the average stroke speed of a motor vehicle and discriminant parameter, determines according to under type:
S54: judge
whether set up, if set up, then show that this road section traffic volume is unimpeded; If be false, enter discriminating step S55;
S55: judge
whether set up, if set up, then show to walk or drive slowly in this section, namely travel speed bends down than unimpeded state, but than high under Traffic Congestion; If be false, enter discriminating step S56;
S56: judge
whether set up, if set up, then show that crowded character is now incidental congestion, if be false, then show to be now recurrent congestion.
Embodiment 2
The difference of the present embodiment and embodiment 1 is only:
Standard deviation algorithm, two exponential smoothing algorithm and MacMaster algorithm etc. can also be adopted based on the abnormality identification of vehicle checker data in the present embodiment.
Three algorithm parameter value K in algorithm
1, K
2, K
3, adopt experience to demarcate, by demarcating after test of many times.
Three algorithm parameter value K in algorithm
1', K
2', K
3', adopt experience to demarcate, by demarcating after test of many times.
Set up example recognition unit as following table 1:
Table 1 is algorithm voting fusion method recognition unit composition
The present embodiment, when merging the ACI algorithm recognition result of originating based on different pieces of information, first will carry out the coupling of time, be unified by the recognition time of two algorithms, thus facilitates blending algorithm to carry out abnormality identification.Data collection cycle for vehicle checker is chosen as 5 minutes, and charge data acquisition mode is Real-time Collection; ACI algorithm based on charge data and the ACI algorithm execution time based on vehicle checker data all shorter.For making the ACI algorithm based on charge data and the ACI algorithm time match based on vehicle checker data, concrete matching principle is as follows:
When carrying out system clock synchronization,
The data collection cycle of vehicle checker is 5 minutes, and therefore the ACI algorithm measuring and calculating cycle of its charge data is also 5 minutes.Charge data acquisition mode is Real-time Collection, and for making system time mate, the ACI algorithm based on charge data chooses 5 minutes equally.
When the ACI algorithm based on charge data and the ACI algorithm recognition time based on vehicle checker data inconsistent time, as: when the recognition time of the ACI algorithm based on charge data is 7:00,7:05,7:10 ... identify the generation of abnormality in 7:05 based on the ACI algorithm of vehicle checker data, then now postpone 5min based on the recognition time of the ACI algorithm of vehicle checker data, namely blending algorithm recognition time is 7:05.
As shown in Figure 4, Fig. 4 is the hardware implementing figure of decision-making voting fusion method system logic, this algorithm fusion adopts the thought of logical circuit " with door " and disjunction gate, first the algorithms of different of same level of trust is differentiated that result is carried out " with door " and merged, after different level of trust is merged after result merged by disjunction gate, finally obtain abnormality recognition result.
What finally illustrate is, above embodiment is only in order to illustrate technical scheme of the present invention and unrestricted, although by referring to the preferred embodiments of the present invention, invention has been described, but those of ordinary skill in the art is to be understood that, various change can be made to it in the form and details, and not depart from the spirit and scope that the present invention limits.
Claims (8)
1., based on the abnormal traffic state recognition methods that vehicle checker and charge data merge, it is characterized in that: comprise the following steps:
S1: gather highway data extraction rate, flow and occupation rate traffic behavior characteristic parameter by vehicle checker;
S2: adopt ACI algorithm identification highway vehicle checker abnormality data according to traffic behavior characteristic parameter;
S3: obtain highway toll data and respective stretch traffic behavior parameter;
S4: calculate road-section average travel speed according to highway toll data and respective stretch traffic behavior parameter;
S5: obtain highway toll abnormality data with comparing of the predetermined average stroke speed of a motor vehicle according to road-section average travel speed;
S6: adopt algorithm voting fusion method according to highway vehicle checker abnormality data and highway toll abnormality data identification highway abnormality data.
2. abnormal traffic state recognition methods of merging based on vehicle checker and charge data according to claim 1, is characterized in that: the traffic behavior characteristic parameter in described step S1 adopts California algorithm, and concrete implementation step is as follows:
S11: obtain upstream and downstream wagon detector data, calculate occupation rate difference and absolute difference:
OCCD=OCC(i,t)-OCC(i+1,t) (1)
In formula, OCCD represents the occupation rate absolute difference of adjacent two vehicle checkers; OCC (i, t) represents upstream vehicle checker occupation rate; OCC (i+1, t) represents downstream vehicle checker occupation rate;
S12: as upstream and downstream difference OCCD < K
1time, can be judged to be now non-blocking up, be normal traffic states; If OCCD>=K
1time, continue next step S13;
Wherein, K
1for the parameter value demarcated by experience;
S13: according to the occupation rate difference between upstream and downstream website, calculates its ratio with upstream stations occupation rate:
S13: as OCCDF < K
2time, can be judged to be now non-blocking up, be normal traffic states; If OCCDF>=K
2time, continue Step4; Wherein, K
2for the parameter value demarcated by experience;
S14: according to the occupation rate difference between upstream and downstream website, calculates its ratio with downstream website occupation rate:
S15: as DOCCDT > K
3time, can be judged to be now non-blocking up, be normal traffic states; If DOCCDT≤K
3time, continue S16; Wherein, K
3for the parameter value demarcated by experience;
S16: read a upper recognition cycle abnormal traffic state, if blocked up, then differentiate that this recognition cycle is that traffic behavior is abnormal, otherwise be normal traffic states, this recognition cycle terminates.
3. abnormal traffic state recognition methods of merging based on vehicle checker and charge data according to claim 1, it is characterized in that: the calculating of the highway toll abnormality data in described step S5, concrete implementation step is as follows:
S51: calculate the average stroke speed of a motor vehicle according to charge data
S52: according to different sections of highway feature, determines discriminant parameter K '
1, K '
2, K '
3value;
S53: determine traffic behavior according to the average stroke speed of a motor vehicle and discriminant parameter, determines according to under type:
S54: judge
whether set up, if set up, then show that this road section traffic volume is unimpeded; If be false, enter discriminating step S55;
S55: judge
whether set up, if set up, then show to walk or drive slowly in this section, namely travel speed bends down than unimpeded state, but than high under Traffic Congestion; If be false, enter discriminating step S56;
S56: judge
whether set up, if set up, then show that crowded character is now incidental congestion, if be false, then show to be now recurrent congestion.
4. abnormal traffic state recognition methods of merging based on vehicle checker and charge data according to claim 1, is characterized in that: the algorithm voting fusion method in described step S6, and concrete implementation step is as follows:
S61: determine that algorithm forms, described algorithm composition comprises the first algorithm and the second algorithm, and described first algorithm is the ACI algorithm based on vehicle checker data; Described second algorithm is the ACI algorithm based on charge data;
S62: the level of trust determining algorithm, continuation test is adopted to define algorithm level of trust, described continuation test is divided into Three Estate: PT=1,2,3, often kind of algorithm all adopts 3 grades of reliability ratings, in continuance test, as PT=1, represent that ACI algorithm provides abnormal traffic state recognition result through a recognition cycle; As PT=2, represent that ACI algorithm provides abnormal traffic state recognition result through two recognition cycles; As PT=3, represent that the duration of continuous 3 or more identifies abnormal traffic state, PT=3 is highest level of trust;
S63: determine recognition unit, described recognition unit is defined as follows:
Determine that the level of trust of the first algorithm is respectively the first level of trust A
1, the second level of trust A
2, the 3rd level of trust A
3;
Determine that the level of trust of the second algorithm is respectively the first level of trust B
1, the second level of trust B
2, the 3rd level of trust B
3;
Determine the first recognition unit A
1b
1, the second recognition unit A
2b
2, the 3rd recognition unit A
3b
3;
Described first recognition unit A
1b
1represent that the first algorithm is the first level of trust A
1be the first level of trust B with the second algorithm
1time the recognition unit determined;
Described second recognition unit A
2b
2represent that the first algorithm is the second level of trust A
2be the second level of trust B with the second algorithm
2time the recognition unit determined;
Described 3rd recognition unit A
3b
3represent that the first algorithm is the 3rd level of trust A
3be the 3rd level of trust B with the second algorithm
3time the recognition unit determined;
S64: the recognition time coupling determining the first algorithm and the second algorithm recognition result, certainty annuity clock synchronous, described system clock synchronization comprise the ACI algoritic module based on charge data system clock, based on the system clock of ACI algoritic module of vehicle checker data and the system clock of Fusion Module; Concrete matching principle is as follows:
The data collection cycle of setting vehicle checker is identical with charge data collection period;
When the ACI algorithm based on charge data is consistent with the ACI algorithm recognition time based on vehicle checker data, the ACI algorithm recognition time set based on charge data is identical with the ACI algorithm recognition time based on vehicle checker data;
When the ACI algorithm based on charge data and the ACI algorithm recognition time based on vehicle checker data inconsistent time, postpone to identify the recognition time of preceding a kind of algorithm.
5., based on the abnormal traffic state recognition system that vehicle checker and charge data merge, it is characterized in that: comprise the ACI algoritic module based on charge data, the ACI algoritic module based on vehicle checker data and Fusion Module;
The described ACI algoritic module based on charge data is used for calculating average travel speed according to charge data traffic behavior parameter and compared with the predetermined average stroke speed of a motor vehicle, thus judges highway toll abnormality data;
The described ACI algoritic module based on vehicle checker data is used for determining highway vehicle checker abnormality data according to the traffic behavior parameter of vehicle checker data acquisition;
Described Fusion Module is used for the highway toll abnormality data obtained by the ACI algoritic module based on charge data and the abnormal traffic state identification signal calculating highway based on the highway vehicle checker abnormality data acquisition algorithm voting fusion method that the ACI algoritic module of vehicle checker data obtains.
6. the abnormal traffic state recognition system merged based on vehicle checker and charge data according to claim 5, is characterized in that: described Fusion Module comprises to be determined algorithm component units, determine algorithm level of trust unit, determines recognition unit and recognition time matching unit;
Describedly determine that algorithm component units is for determining the algorithm ingredient needing to merge, described algorithm composition comprises the first algorithm and the second algorithm, and described first algorithm is the ACI algorithm based on vehicle checker data; Described second algorithm is the ACI algorithm based on charge data;
Describedly determine that algorithm level of trust unit is for determining the level of trust of algorithm ingredient, described algorithm level of trust adopts continuation test to define algorithm level of trust, described continuation test is divided into Three Estate: PT=1,2,3, often kind of algorithm all adopts 3 grades of reliability ratings, in continuance test, as PT=1, represent that ACI algorithm provides abnormal traffic state recognition result through a recognition cycle; As PT=2, represent that ACI algorithm provides abnormal traffic state recognition result through two recognition cycles; As PT=3, represent that the duration of continuous 3 or more identifies abnormal traffic state, PT=3 is highest level of trust;
Describedly determine that recognition unit is for determining according to algorithm level of trust and algorithm ingredient the recognition result that algorithm forms, described recognition unit is defined as follows:
Determine that the level of trust of the first algorithm is respectively the first level of trust A
1, the second level of trust A
2, the 3rd level of trust A
3;
Determine that the level of trust of the second algorithm is respectively the first level of trust B
1, the second level of trust B
2, the 3rd level of trust B
3;
Determine the first recognition unit A
1b
1, the second recognition unit A
2b
2, the 3rd recognition unit A
3b
3;
Described first recognition unit A
1b
1represent that the first algorithm is the first level of trust A
1be the first level of trust B with the second algorithm
1time the recognition unit determined;
Described second recognition unit A
2b
2represent that the first algorithm is the second level of trust A
2be the second level of trust B with the second algorithm
2time the recognition unit determined;
Described 3rd recognition unit A
3b
3represent that the first algorithm is the 3rd level of trust A
3be the 3rd level of trust B with the second algorithm
3time the recognition unit determined;
Described recognition time matching unit is used for the recognition time of unified algorithms of different composition, for determining the recognition time coupling of the first algorithm and the second algorithm recognition result, certainty annuity clock synchronous, described system clock synchronization comprise the ACI algoritic module based on charge data system clock, based on the system clock of ACI algoritic module of vehicle checker data and the system clock of Fusion Module; Concrete matching principle is as follows:
The data collection cycle of setting vehicle checker is identical with charge data collection period;
When the ACI algorithm based on charge data is consistent with the ACI algorithm recognition time based on vehicle checker data, the ACI algorithm recognition time set based on charge data is identical with the ACI algorithm recognition time based on vehicle checker data;
When the ACI algorithm based on charge data and the ACI algorithm recognition time based on vehicle checker data inconsistent time, postpone to identify the recognition time of preceding a kind of algorithm.
7. the abnormal traffic state recognition system merged based on vehicle checker and charge data according to claim 5, it is characterized in that: described acquisition based on the traffic behavior characteristic parameter in the ACI algoritic module of vehicle checker data adopts California algorithm, and concrete implementation step is as follows:
S11: obtain upstream and downstream wagon detector data, calculate occupation rate difference (absolute difference):
OCCD=OCC(i,t)-OCC(i+1,t) (1)
In formula, OCCD represents the occupation rate absolute difference of adjacent two vehicle checkers; OCC (i, t) represents upstream vehicle checker occupation rate; OCC (i+1, t) represents downstream vehicle checker occupation rate;
S12: as upstream and downstream difference OCCD < K
1time, can be judged to be now non-blocking up, be normal traffic states; If OCCD>=K
1time, continue next step S13;
Wherein, K
1, K
2, K
3for the parameter value demarcated by experience;
S13: according to the occupation rate difference between upstream and downstream website, calculates its ratio with upstream stations occupation rate:
S13: as OCCDF < K
2time, can be judged to be now non-blocking up, be normal traffic states; If OCCDF>=K
2time, continue Step4; Wherein, K
1, K
2, K
3for the parameter value demarcated by experience;
S14: according to the occupation rate difference between upstream and downstream website, calculates its ratio with downstream website occupation rate:
S15: as DOCCDT > K
3time, can be judged to be now non-blocking up, be normal traffic states; If DOCCDT≤K
3time, continue S16; Wherein, K
3for the parameter value demarcated by experience;
S16: read a upper recognition cycle abnormal traffic state, if blocked up, then differentiate that this recognition cycle is that traffic behavior is abnormal, otherwise be normal traffic states, this recognition cycle terminates.
8. the abnormal traffic state recognition system merged based on vehicle checker and charge data according to claim 5, it is characterized in that: the described calculating based on the highway toll abnormality data in the ACI algoritic module of charge data, concrete implementation step is as follows:
S51: calculate the average stroke speed of a motor vehicle according to charge data
S52: according to different sections of highway feature, determines discriminant parameter K '
1, K '
2, K '
3value;
S53: determine traffic behavior according to the average stroke speed of a motor vehicle and discriminant parameter, determines according to under type:
S54: judge
whether set up, if set up, then show that this road section traffic volume is unimpeded; If be false, enter discriminating step S55;
S55: judge
whether set up, if set up, then show to walk or drive slowly in this section, namely travel speed bends down than unimpeded state, but than high under Traffic Congestion; If be false, enter discriminating step S56;
S56: judge
whether set up, if set up, then show that crowded character is now incidental congestion, if be false, then show to be now recurrent congestion.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201410604146.7A CN104361349A (en) | 2014-10-31 | 2014-10-31 | Car inspection device and toll data fusion based abnormal traffic state identification method and system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201410604146.7A CN104361349A (en) | 2014-10-31 | 2014-10-31 | Car inspection device and toll data fusion based abnormal traffic state identification method and system |
Publications (1)
Publication Number | Publication Date |
---|---|
CN104361349A true CN104361349A (en) | 2015-02-18 |
Family
ID=52528607
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201410604146.7A Pending CN104361349A (en) | 2014-10-31 | 2014-10-31 | Car inspection device and toll data fusion based abnormal traffic state identification method and system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN104361349A (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106297285A (en) * | 2016-08-17 | 2017-01-04 | 重庆大学 | Freeway traffic running status fuzzy synthetic appraisement method based on changeable weight |
CN110197586A (en) * | 2019-05-20 | 2019-09-03 | 重庆大学 | A kind of express highway section congestion detection method based on multi-source data |
CN111882858A (en) * | 2020-06-01 | 2020-11-03 | 重庆大学 | Method for predicting queuing length of expressway abnormal events based on multi-source data |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20070005229A1 (en) * | 2003-12-19 | 2007-01-04 | Bayerische Motoren Werke | Verifying the validity of traffic status information |
CN101571997A (en) * | 2009-05-31 | 2009-11-04 | 上海宝康电子控制工程有限公司 | Method and device for fusion processing of multi-source traffic information |
CN104021671A (en) * | 2014-05-16 | 2014-09-03 | 浙江银江研究院有限公司 | Real-time road condition determination method through combined svm and fuzzy determination mode |
-
2014
- 2014-10-31 CN CN201410604146.7A patent/CN104361349A/en active Pending
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20070005229A1 (en) * | 2003-12-19 | 2007-01-04 | Bayerische Motoren Werke | Verifying the validity of traffic status information |
CN101571997A (en) * | 2009-05-31 | 2009-11-04 | 上海宝康电子控制工程有限公司 | Method and device for fusion processing of multi-source traffic information |
CN104021671A (en) * | 2014-05-16 | 2014-09-03 | 浙江银江研究院有限公司 | Real-time road condition determination method through combined svm and fuzzy determination mode |
Non-Patent Citations (1)
Title |
---|
韩坤林: "基于车检器及收费数据融合的高速公路异常状态识别研究", 《万方数据库》 * |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106297285A (en) * | 2016-08-17 | 2017-01-04 | 重庆大学 | Freeway traffic running status fuzzy synthetic appraisement method based on changeable weight |
CN106297285B (en) * | 2016-08-17 | 2018-09-21 | 重庆大学 | Freeway traffic operating status fuzzy synthetic appraisement method based on changeable weight |
CN110197586A (en) * | 2019-05-20 | 2019-09-03 | 重庆大学 | A kind of express highway section congestion detection method based on multi-source data |
CN111882858A (en) * | 2020-06-01 | 2020-11-03 | 重庆大学 | Method for predicting queuing length of expressway abnormal events based on multi-source data |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110533912B (en) | Driving behavior detection method and device based on block chain | |
CN103886756B (en) | Based on the freeway network method for detecting operation state of OBU | |
CN103839409B (en) | Based on the traffic flow modes method of discrimination of multibreak facial vision sensing cluster analysis | |
CN100454354C (en) | Road traffic accident multi-happening section identifying method | |
CN104392607B (en) | A kind of traffic state judging method and device | |
CN103903433A (en) | Real-time dynamic judgment method and device for road traffic state | |
CN101571997A (en) | Method and device for fusion processing of multi-source traffic information | |
CN103077610A (en) | Road trip time estimating method and system | |
CN103150900A (en) | Traffic jam event automatic detecting method based on videos | |
US20220383738A1 (en) | Method for short-term traffic risk prediction of road sections using roadside observation data | |
CN109841078B (en) | Navigation data processing method and device and storage medium | |
CN107885795A (en) | A kind of data verification method, system and the device of bayonet socket data | |
CN103678834A (en) | Modeling method and modeling device for cyber-physical system (CPS) | |
CN109003442B (en) | Road delay time calculation and traffic jam situation determination method and system | |
CN104700630A (en) | Method and system for monitoring vehicle flow of highway | |
CN104298832A (en) | Road network traffic flow analytical method based on RFID technology | |
WO2024060588A1 (en) | Expressway tunnel accident real-time risk prediction method and apparatus, device, and medium | |
CN104361349A (en) | Car inspection device and toll data fusion based abnormal traffic state identification method and system | |
Evans et al. | Evolution and future of urban road incident detection algorithms | |
Huang et al. | Real time delay estimation for signalized intersection using transit vehicle positioning data | |
Zhang et al. | An approach for real-time urban traffic state estimation by fusing multisource traffic data | |
Hale et al. | Evaluation of data‐driven performance measures for comparing and ranking traffic bottlenecks | |
US20230126364A1 (en) | Systems and signal processing methods for real-time traffic congestion detection | |
CN107134148A (en) | Road occupation for parking behavioral value method and apparatus | |
CN103280109A (en) | Obtaining method, obtaining device and prediction system of travel time |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20150218 |