CN102708688B - Secondary fuzzy comprehensive discrimination-based urban road condition recognition method - Google Patents

Abstract

An urban road state identification method based on two-level fuzzy comprehensive discrimination involves intelligent traffic control. It includes several steps of automatic division of urban road sub-sections, real-time reception of road traffic parameters, real-time identification of road traffic status and real-time dynamic display of road traffic guidance screens. In the urban road sub-section automatic division module, a road is automatically divided into two sub-sections. The sub-section division depends on parameters including road length, green signal ratio of front signal lights, road design saturation rate, road speed limit and control parameters. Real-time identification of urban road traffic status is carried out in the second-level fuzzy comprehensive urban road traffic status identification module. This identification is based on the second-level fuzzy identification based on the theory of segmented Gaussian membership function and fuzzy mathematical theory. After the comprehensive evaluation of the second-level fuzzy, the result of urban road state identification is obtained and published on the road traffic guidance screen. The invention fully considers the actual factors of urban roads, effectively realizes traffic load balance, and has broad application prospects.

Description

An urban road state identification method based on two-level fuzzy comprehensive discrimination

technical field

The invention belongs to the field of computer application technology, in particular to intelligent traffic control.

Background technique

The state of road traffic can be divided into three states: "jam", "crowded" and "smooth". The purpose of road traffic state evaluation is to use certain methods to analyze real-time traffic data, quickly identify various traffic states on the road, and provide a basis for traffic control and guidance.

The identification methods of road traffic status can be divided into two categories: artificial identification methods and automatic identification methods. The former includes citizen reports, specialist reports and closed-circuit television surveillance. The advantage of this method is that it is convenient, direct, and economical. The disadvantage is that it requires witnesses at the time and place, and it is difficult to play a role 24 hours a day. The latter is based on information collection and processing technology, computer technology and communication technology, which can continuously monitor the state of road traffic around the clock, and has received more and more attention and considerable development.

Most of the existing traffic state identification methods at home and abroad are based on the sudden traffic incidents of expressways. In urban traffic, due to the influence of intersection lights and non-motorized vehicles, the traffic flow characteristics are similar to those of expressways. The comparison is more complicated, and it is more difficult to automatically judge the urban traffic state. The data processing methods used in urban traffic status mainly include conventional methods such as decision tree, statistical analysis, and smoothing filter. However, the change of parameter index based on the evolution process of the traffic flow state itself is a continuous process, and the division between various states is also fuzzy. The method of fuzzy discrimination is more suitable for judging the traffic flow state. In the prior art, fuzzy discrimination generally adopts first-level discrimination, and the first-level fuzzy discrimination method is based on the traffic characteristic parameters of the entire road following the criterion of the maximum degree of membership to determine the traffic state. This method has the following disadvantages:

First, it does not consider the influence of traffic characteristic parameters on urban roads near traffic lights and far away from signal lights on the traffic state of corresponding roads. These traffic characteristic parameters include queuing length, average speed and lane occupancy, etc. The accuracy of the traffic status obtained by the existing first-level fuzzy discrimination method needs to be further verified;

Second, in the selection of membership functions in fuzzy traffic state discrimination, in order to simplify, most of them use the reduced half-trapezoidal formula to represent linearly, ignoring the fact that Gaussian functions have no zero-crossing points, smooth curves, and clear physical meanings, which are more suitable as fuzzy traffic conditions. The characteristics of the membership function of the discriminant, there are less methods for determining the form and parameters of the membership function of the traffic characteristic parameters in the adjacent signal light road section;

Third, due to the complexity of urban road traffic congestion, each traffic state has a certain similarity, which makes the division of traffic states ambiguous. For cases where the differences in all membership degrees are not obvious, only the principle of maximum membership degree To give the final judgment result cannot describe the objective ambiguity of the traffic state.

Based on the above reasons, the existing traffic state identification methods cannot objectively and accurately distinguish the urban road state.

Contents of the invention

The purpose of the present invention is to propose a method for urban road traffic state identification based on two-level fuzzy comprehensive discrimination to address the shortcomings of traditional traffic state identification methods. The urban road traffic state identification method proposed by the present invention fully considers the traffic reality of urban roads, and constructs a mathematical model for automatic subsections of urban roads; The Gaussian membership function of the characteristics of the discriminant membership function, and then adopt the second-level fuzzy comprehensive discrimination; finally realize the more objective and accurate identification of urban road traffic status.

The purpose of the present invention is achieved like this: urban road state discrimination comprises several steps of urban road sub-section automatic division, real-time receiving road traffic parameters, urban road traffic state real-time identification and road traffic guidance screen real-time dynamic display; urban road sub-section automatically The division is carried out in the urban road sub-section automatic division module, the real-time reception of road traffic parameters is carried out in the urban road sub-section traffic parameter acquisition module, and the real-time identification of road traffic status is carried out in the secondary fuzzy comprehensive urban road traffic state identification module; The subsection automatic division module, the urban road subsection traffic parameter collection and the secondary fuzzy comprehensive urban road traffic state identification module are set in the processing server of the intelligent traffic control system.

The urban road sub-section automatic division module receives the urban road design parameter module parameters from the intelligent traffic control system, completes the urban road sub-section division and sends the division result to the second-level fuzzy comprehensive urban road traffic status identification module, and the second-level fuzzy comprehensive The urban road traffic state identification module simultaneously receives the video parameters and the urban road sub-section division result parameters transmitted by the urban road sub-section traffic parameter acquisition module through the network in real time from the road site collection point to identify the urban road traffic state in real time, and the real-time traffic state identification results Publish to the road traffic guidance screen of the intelligent traffic control system through network communication, and the road traffic guidance screen dynamically releases the traffic status identification results in real time.

The automatic division of urban road sub-sections is to automatically divide a road U into two sub-sections U1 and U2, and the division of U1 and U2 depends on parameters including road length, green signal ratio of signal lights ahead, road design saturation rate, road speed limit and control parameters .

The real-time identification of road traffic status is a secondary fuzzy identification based on the theory of segmented Gaussian membership function and fuzzy mathematics theory, the steps are:

(1) Establish the evaluation object single factor set U _i for each road sub-section;

(2) Establish an evaluation set F _i for each road subsection;

(3) Establish a fuzzy relationship mapping from the single factor set U _i to the evaluation set F _i , and derive the single factor evaluation matrix R _i from the Cartesian product correspondence

(4) For the first-level fuzzy comprehensive evaluation, select the segmented Gaussian fuzzy mathematical comprehensive function for comprehensive and normalized processing;

(5) Second-level fuzzy comprehensive evaluation;

(6) Carry out fuzzy analysis and judgment on the results of the secondary judgment, and obtain the result of urban road status identification.

The automatic division of the urban road subsections automatically divides a road U into two subsections U1 and U2 according to formula 1-1:

$\left\{\begin{array}{c}{\mathrm{<span\; class="notranslate">\; d</span>}}_{\mathrm{<span\; class="notranslate">\; u</span>}\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; f</span>}<span\; class="notranslate">\; (</span>\stackrel{<span\; class="notranslate">\; \&OverBar;</span>}{\mathrm{<span\; class="notranslate">\; d</span>}}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; t</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; the\; s</span>}<span\; class="notranslate">\; ,</span>\stackrel{<span\; class="notranslate">\; \&OverBar;</span>}{\mathrm{<span\; class="notranslate">\; v</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\stackrel{<span\; class="notranslate">\; \&OverBar;</span>}{\mathrm{<span\; class="notranslate">\; d</span>}}<span\; class="notranslate">\; /</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; a</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; t</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; b</span>}<span\; class="notranslate">\; /</span>\stackrel{<span\; class="notranslate">\; \&OverBar;</span>}{\mathrm{<span\; class="notranslate">\; v</span>}}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; the\; s</span>}<span\; class="notranslate">\; )</span>\\ {\mathrm{<span\; class="notranslate">\; d</span>}}_{\mathrm{<span\; class="notranslate">\; u</span>}\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; =</span>\stackrel{<span\; class="notranslate">\; \&OverBar;</span>}{\mathrm{<span\; class="notranslate">\; d</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; d</span>}}_{\mathrm{<span\; class="notranslate">\; u</span>}\mathrm{<span\; class="notranslate">\; 1</span>}}\end{array}\right.<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; )</span>$

表示整条道路长度、t表示路段前方信号灯绿信比、s表示道路设计饱和率、

表示道路限制速度，a是与道路总长相关的控制参数，b是与道路限制速度相关的控制参数。In the formula, d _U1 and d _U2 represent the lengths of road subsections U1 and U2 respectively,

Indicates the length of the entire road, t indicates the green signal ratio of the signal lights ahead of the road section, s indicates the saturation rate of the road design,

Indicates the road speed limit, a is the control parameter related to the total length of the road, and b is the control parameter related to the road speed limit.

The specific steps of the automatic identification of the two-level fuzzy comprehensive urban road traffic state are:

(1) Establish a single-factor set U _i of evaluation objects for each road sub-section; U _i = [L, V, D], where i=1, 2 indicates the sub-set of the road section, and L represents the road queue length ratio, V represents the average vehicle speed, D represents the occupancy rate;

(2) Establish an evaluation set F _i for each road subsection; F _i =[f _i1 , f _i2 , f _i3 ], where f _i1 represents that the i-th sub-section of the road belongs to the smooth state, and f _i2 represents that the i-th sub-section belongs to Congested state, f _i3 represents that the i-th sub-section belongs to the congested state;

(3) Establish a fuzzy relationship mapping from the single factor set U _i to the evaluation set F _i , so that any element u in the factor set is the Cartesian product of L, V, D L×V×D={( l,v,d)|l∈L,v∈V,d∈D}, the corresponding elements (l,v,d) are uniquely corresponding, thus deriving the single-factor evaluation matrix R _i , R _i =[R _i1 , R _i2 ,R _i3 ] ^T , where R _i1 refers to the degree to which the queue length ratio of sub-section i belongs to smoothness, congestion and congestion, R _i2 refers to the degree to which the speed of sub-section i belongs to smoothness, congestion and congestion, and R _i3 refers to the degree that the occupancy rate of sub-road section i belongs to smooth flow, congestion and congestion, where l represents the value of the queue length ratio variable, v represents the variable value of the average vehicle speed, and d represents the variable value of the occupancy rate;

(4) For the first-level fuzzy comprehensive evaluation, select the appropriate segmental Gaussian fuzzy mathematical _{comprehensive} function for comprehensive, and use the corresponding fuzzy set A _i =[a _i1 ,a _i2 , a _i3 ] represents the weight distribution of the factor,

Among them, i can take the serial number 1 or 2 of the sub-sections divided into sub-sections, a _i1 , a _i2 , a _i3 represent the proportions of the queuing length ratio, average vehicle speed and occupancy rate of the i-th sub-section respectively in the fuzzy evaluation, and obtain a Level single-factor comprehensive evaluation set B _i =[b _i1 ,b _i2 ,b _i3 ]=A _i οR _i , and normalize it, where b _i1 , b _i2 , b _i3 respectively indicate the i-th sub-section belongs to To the degree of unimpeded, crowded and blocked state;

(5) Second-level fuzzy comprehensive evaluation, using the output of the previous level of evaluation as the evaluation matrix R ^～ =[B ₁ , B ₂ ] ^T , where B ₁ and B ₂ respectively represent the first sub-section and the second sub-section calculated in the previous step. For the first-level single-factor comprehensive evaluation set of sub-sections, the weight fuzzy subset of each sub-section to the whole road is A ^～ , then the second-level fuzzy evaluation output B ^～ =A ^～ οR ^～ =[b ₁ ^～ ,b ₂ ^～ , b ₃ ^～ ], where b ₁ ^～ , b ₂ ^～ , b ₃ ^～ represent respectively the degree of traffic status of the whole road section after the second-level fuzzy evaluation;

(6) Carry out fuzzy analysis and judgment on the results of the second-level judgment, and obtain the result of urban road state identification: set a threshold λ∈[0,1], for any b _j ^～ ≥ λ(j=1,2,3 ) all meet the requirements, where the values 1, 2, and 3 that can be taken by the parameter j correspond to the status signs of smooth traffic, congestion, and congestion of the entire road section, and the parameter b _j ^～ indicates that the traffic state of the entire road section belongs to the corresponding logo j after the second-level fuzzy evaluation degree of traffic status. When only one value of b _j ^～ is greater than λ, it is normalized to the corresponding traffic state; when the values of b ₁ ^～ and b ₂ ^～ are all greater than λ, it is normalized to the “smooth/congested” critical state; when the values of b ₂ ^~ , b ₃ ^~ are all greater than λ, they are normalized to the critical state of "congestion/blockage".

The present invention has the following advantages:

(1) Fully consider the influence of actual factors of urban roads, realize more objective and accurate identification of urban road traffic status, and provide more effective information support for dynamic control and guidance of urban intelligent traffic. In this way, the load balance of urban road traffic can be achieved, congestion can be alleviated, traffic can be smooth and orderly, and a good foundation can be laid for the harmony of urban road traffic.

(2) The identification algorithm has high computational efficiency and broad application prospects.

Description of drawings

Fig. 1 is a schematic diagram of the structure of the present invention in the intelligent traffic control system.

Fig. 2 is a schematic diagram of the segmental Gaussian membership function of the second sub-segment to the average vehicle queuing length ratio in the embodiment.

Fig. 3 is a schematic diagram of the Gaussian membership function of the second sub-section to the average speed in the embodiment.

Fig. 4 is a schematic diagram of the segmental Gaussian membership function of the second sub-section to the lane occupancy rate in the embodiment.

Fig. 5 is a dynamic display diagram of the road traffic guidance screen. In the figure, the solid line represents the red color in the actual guidance screen, indicating a congested traffic state; the dotted line represents the yellow color in the actual guidance screen, indicating a congested traffic state; the dotted line represents the green color in the actual guidance screen, indicating a smooth traffic state .

Detailed ways

See attached drawing 1. The urban road sub-section automatic division module and the second-level fuzzy comprehensive urban road traffic state identification module are the core contents of the present invention, which are completed by a processing server. The urban road design parameter input module is based on the planning and design of each urban road, which corresponds to the entire road length, the green signal ratio of the signal lights in front of the road section, the saturation rate of the road design, and the road speed limit, etc., and forms the configuration according to the actual urban road traffic network The files are stored on the processing server. The urban road sub-section automatic division module divides the urban road into sub-sections according to the urban road design parameters combined with the automatic division mathematical model. The collection of traffic parameters of urban road sub-sections is realized based on the existing mature video collection and analysis technology, and is transmitted from the road site collection point to the processing server in real time through the network. The second-level fuzzy comprehensive urban road traffic status identification module is realized by using piecewise Gaussian membership function and fuzzy mathematical theory. The road traffic guidance screen dynamically releases the traffic status module to receive real-time traffic status identification results through network communication, and is used to realize traffic guidance at key sites of actual urban road traffic.

In the specific implementation process, the identification of the road status of the whole city includes several steps such as automatic division of road subsections, real-time reception of road traffic parameters, real-time identification of road traffic status and real-time dynamic display of road traffic guidance screens.

Step 1: Automatic division of road subsections

In this embodiment, the parameters of a certain urban road design are set: the maximum speed limit is 40 km/h, the traffic light cycle is 60 seconds and the green signal ratio is 1/2, the total length of the road is 1 km, the road saturation rate is 0.4, and Set the control parameters a and b in formula 1-1 as 3 and 10 km/h respectively, according to formula 1-1

$\left\{\begin{array}{c}{\mathrm{<span\; class="notranslate">\; d</span>}}_{\mathrm{<span\; class="notranslate">\; u</span>}\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; f</span>}<span\; class="notranslate">\; (</span>\stackrel{<span\; class="notranslate">\; \&OverBar;</span>}{\mathrm{<span\; class="notranslate">\; d</span>}}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; t</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; the\; s</span>}<span\; class="notranslate">\; ,</span>\stackrel{<span\; class="notranslate">\; \&OverBar;</span>}{\mathrm{<span\; class="notranslate">\; v</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\stackrel{<span\; class="notranslate">\; \&OverBar;</span>}{\mathrm{<span\; class="notranslate">\; d</span>}}<span\; class="notranslate">\; /</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; a</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; t</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; b</span>}<span\; class="notranslate">\; /</span>\stackrel{<span\; class="notranslate">\; \&OverBar;</span>}{\mathrm{<span\; class="notranslate">\; v</span>}}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; the\; s</span>}<span\; class="notranslate">\; )</span>\\ {\mathrm{<span\; class="notranslate">\; d</span>}}_{\mathrm{<span\; class="notranslate">\; u</span>}\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; =</span>\stackrel{<span\; class="notranslate">\; \&OverBar;</span>}{\mathrm{<span\; class="notranslate">\; d</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; d</span>}}_{\mathrm{<span\; class="notranslate">\; u</span>}\mathrm{<span\; class="notranslate">\; 1</span>}}\end{array}\right.<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; )</span>$

The calculated lengths of the road sub-sections are: d _U1 = 0.32 km and d _U2 = 0.68 km.

Step 2: Receive road traffic parameters in real time

The road traffic parameter acquisition method is to set up a camera in front of the road to obtain real-time road video, calibrate the position in the video according to the sub-section of the road, and use the existing video analysis and processing technology to obtain real-time traffic parameters of each sub-section. In this embodiment, the vehicle queuing length ratio, average vehicle speed, and lane occupancy rate averaged over 5 minutes are taken.

Step 3: Real-time identification of urban road traffic status

In the process of real-time identification of road traffic status, follow the following 6 steps:

(1) Establish a single-factor set U _i of evaluation objects for each road sub-section; U _i = [L, V, D], where i=1, 2 represents the sub-set of the road segment, L represents the queue length ratio, V Represents the average vehicle speed, and D represents the occupancy rate.

(2) Establish an evaluation set F _i for each sub-section. F _i =[f _i1 ,f _i2 ,f _i3 ], where f _i1 represents that the i-th sub-section of the road is in a smooth state, f _i2 represents that the i-th sub-section is in a congested state, and f _i3 represents that the i-th sub-section is in a congested state.

(3) Establish a single-factor evaluation, that is, establish a fuzzy relational mapping from the single-factor set U _i to the sub-evaluation set F _i , so that any element u in the factor set is multiplied by the Cartesian product of L, V, and D The corresponding elements (l, v, d) in L×V×D={(l,v,d)|l∈L, v∈V, d∈D} are uniquely corresponding, so the single-factor evaluation matrix R can be derived _i ，R _i =[R _i1 ,R _i2 ,R _i3 ] ^T , where R _i1 refers to the degree of queue length ratio of sub-section i belonging to smooth flow, congestion and congestion, and R _i2 refers to the degree of sub-section i speed belonging to The degree of smoothness, congestion and congestion, R _i3 refers to the degree to which the occupancy rate of sub-section i belongs to smoothness, congestion and congestion.

(4) For the first-level fuzzy comprehensive evaluation, select the appropriate segmental Gaussian fuzzy mathematical _{comprehensive} function for comprehensive, and use the corresponding fuzzy set A _i =[a _i1 ,a _i2 , a _i3 ] represents the weight distribution of this factor, and the first-level single-factor comprehensive evaluation set B _i =[b _i1 ,b _i2 ,b _i3 ]=A _i οR _i is obtained and normalized.

(5) Second-level fuzzy comprehensive evaluation, taking the output of the previous level of evaluation as the evaluation matrix R ^～ =[B ₁ ,B ₂ ] ^T , and setting the fuzzy subset of the weight of each sub-section to the entire road as A ^～ , then it can be obtained The secondary fuzzy evaluation output B ^～ =A ^～ οR ^～ =[b ₁ ^～ ,b ₂ ^～ ,b ₃ ^～ ].

(6) Carry out fuzzy analysis and judgment on the second-level judgment results, set a threshold λ∈[0,1], and meet the requirements for any b _j ^～ ≥ λ(j=1,2,3), when b _j ^～ When there is only one value greater than λ, it is normalized to the corresponding traffic state; when the values of b ₁ ^～ , b _{2 ～} ^are all greater than λ, it is normalized to the critical state of “smooth/congested”; when b When the values of ₂ ^～ , b ₃ ^～ are all greater than λ, they will be normalized to the critical state of "congestion/blockage". For example, the second-level judgment result is (0.4, 0.4, 0.2). If λ is 0.3, the final result judgment The road traffic status is "smooth/congested".

The segmental Gaussian membership functions of the average vehicle queuing length ratio, average speed, and lane occupancy rate of the second sub-section of the present embodiment are shown in Figures 2, 3, and 4.

Step 4: Real-time dynamic display on the road traffic guidance screen

The road traffic conditions obtained after the fuzzy analysis and judgment of the secondary judgment results are transmitted to the road traffic guidance screen for real-time display, and are divided into three colors: red, yellow, and green to show congestion, congestion, and smooth conditions respectively.

Accompanying drawing 5 has provided the real-time dynamic display of the road traffic guidance screen of the present embodiment. In the actual traffic guidance screen, the red color represents the congested traffic state, the yellow color represents the congested traffic state, and the green color represents the unimpeded specific situation. Accompanying drawing represents the red in the traffic guidance screen with the solid line, the yellow in the traffic guidance screen is represented by the dotted line, and the green in the traffic guidance screen is represented by the dotted line. The road traffic conditions are clear at a glance, which provides great convenience for road traffic participants.

Claims (3)

1. the urban road state of differentiating based on Two-stage Fuzzy Comprehensive is distinguished a method for distinguishing, it is characterized in that: distinguish that method for distinguishing comprises that urban road subsegment divides, receives in real time road traffic parameter, urban road traffic state real-time identification and the several steps of road traffic induced screen Real time dynamic display automatically; Urban road subsegment is automatically divided in urban road subsegment and automatically divides in module and carry out, receive in real time road traffic parameter and carry out in urban road subsegment traffic parameter acquisition module, urban road traffic state real-time identification is carried out in Two-stage Fuzzy Comprehensive urban road traffic state recognition module; Urban road subsegment is automatically divided module, the traffic parameter collection of urban road subsegment and Two-stage Fuzzy Comprehensive urban road traffic state recognition module and is arranged in the processing server of intelligent traffic control system;

Urban road subsegment is automatically divided module and is received the parameter from the designing of city road parameter module in intelligent traffic control system, complete urban road subsegment is divided and division result is sent to Two-stage Fuzzy Comprehensive urban road traffic state recognition module, Two-stage Fuzzy Comprehensive urban road traffic state recognition module receives urban road subsegment traffic parameter acquisition module simultaneously and from video parameter and the urban road subsegment division result parameter of road collection in worksite point real-time Transmission, carries out urban road traffic state real-time identification by network, real-time traffic states identification result is published to the road traffic induced screen of intelligent traffic control system by network communication, the in real time dynamic issuing traffic state identification result of road traffic induced screen,

It is that a road U is divided into two subsegment U1 and U2 automatically that described urban road subsegment is divided automatically, and U1 and U2 divide and depend on that parameter comprises link length, front signal lamp split, highway layout saturation factor, road limits speed and controls parameter;

Described real-time identification road traffic state is being fuzzy the distinguishing of secondary of carrying out based on segmentation Gauss member function and fuzzy mathematics theory, and step is:

(1), each road subsegment is set up to evaluation object list set of factors U _i;

(2), for each road subsegment, set up evaluation collection F _i;

(3), set up from single set of factors U _ito evaluation collection F _ithe mapping of fuzzy relation, by Cartesian product corresponding relation, derive single factor evaluation matrix R _i;

(4), first order Fuzzy Synthetic Evaluation, select segmentation Gaussian Blur mathematical synthesis function carry out comprehensive and made normalized;

(5) Two-stage Fuzzy Comprehensive Evaluation;

(6), secondary result of determination is carried out to fuzzy analysis judgement, draw the result that urban road state is distinguished.

2. urban road state as claimed in claim 1 is distinguished method for distinguishing, it is characterized in that:

A road U is divided into two subsegment U1 to the automatic division of described urban road subsegment automatically and U2 divides according to formula 1-1:

$\left\{\begin{array}{c}{d}_{U1}=f(\stackrel{\&OverBar;}{d},t,s,\stackrel{\&OverBar;}{v})=\stackrel{\&OverBar;}{d}/(a-t+b/\stackrel{\&OverBar;}{v}+s)\\ d_{U2}=\stackrel{\&OverBar;}{d}-d_{U1}\end{array}\right.---(1-1)$

In formula, d _u1and d _u2the length that represents respectively road subsegment U1 and U2,

represent entire road length, t represent section front signal lamp split, s represent highway layout saturation factor,

represent road limits speed, a is the control parameter with the total long correlation of road, and b is the control parameter with road limits velocity correlation.

3. urban road state as claimed in claim 1 is distinguished method for distinguishing, it is characterized in that: the concrete steps of described Two-stage Fuzzy Comprehensive urban road traffic state automatic Identification are:

(1), each road subsegment is set up to evaluation object list set of factors U _i; U _i=[L, V, D], i=1 wherein, which section subsets 2 represent, and L represents section queue length ratio, and V represents average speed, and D represents occupancy;

(2), for each road subsegment, set up evaluation collection F _i; F _i=[f _i1, f _i2, f _i3], f wherein _i1represent that road i sub-section belongs to unimpeded state, f _i2represent that i sub-section belongs to congestion state, f _i3represent that i sub-section belongs to blocked state;

(3), set up from single set of factors U _ito evaluation collection F _ithe mapping of fuzzy relation, single set of factors U _iin arbitrary element u just and L, V, Cartesian product L * V * D={ (l, v, d) of D | l ∈ L, v ∈ V, the unique correspondence of corresponding element (l, v, d) in d ∈ D}, derives single factor evaluation matrix R thus _i, R _i=[R _i1, R _i2, R _i3] ^t, R wherein _i1refer to the degree that the i queue length of sub-section is more unimpeded than being under the jurisdiction of, crowded and stop up, R _i2refer to sub-section i speed and be under the jurisdiction of degree unimpeded, crowded and that stop up, R _i3refer to sub-section i occupancy and be under the jurisdiction of degree unimpeded, crowded and that stop up, l represents that queue length represents that than variable-value, v average speed variable-value, d represent occupancy variable-value;

(4), first order Fuzzy Synthetic Evaluation, according to expertise and actual observation statistics, select the segmentation Gaussian Blur mathematical synthesis function of each traffic characteristic amount to carry out comprehensively, to each sub-section single set of factors U for collection _iinterior corresponding fuzzy set A _i=[a _i1, a _i2, a _i3] represent the weight allocation of this factor, wherein, the sequence number 1 or 2 in i section that desirable section is divided into, a _i1, a _i2, a _i3represent respectively the sub-section of i queue length ratio, average speed, occupancy shared proportion in fuzzy assessment, obtain one-level list combined factors evaluation collection B _i=[b _i1, b _i2, b _i3]=A _iο R _i, and made normalized, wherein b _i1, b _i2, b _i3represent that respectively the sub-section of i is under the jurisdiction of degree unimpeded, crowded, blocked state;

(5) Two-stage Fuzzy Comprehensive Evaluation, using previous stage evaluation output as evaluation matrix R ^~=[B ₁, B ₂] ^t, B wherein ₁, B ₂the one-level list combined factors evaluation collection that represents respectively the 1st sub-section that previous step is obtained and the 2nd sub-section is A by each sub-section to the weight fuzzy subset of whole road ^~, can obtain secondary fuzzy assessment output B ^~=A ^~ο R ^~=[b ₁ ^~, b ₂ ^~, b ₃ ^~], b wherein ₁ ^~, b ₂ ^~, b ₃ ^~after representing respectively secondary fuzzy assessment, whole road section traffic volume state is under the jurisdiction of degree unimpeded, crowded, that stop up;

(6), secondary result of determination is carried out to fuzzy analysis judgement, draw the result that urban road traffic state is distinguished: set a threshold value λ ∈ [0,1], to any b _j ^~>=λ (j=1,2,3) all meets the requirements, and value that parameter j can get 1,2,3 is unimpeded, crowded, the blocked state sign of corresponding whole road section traffic volume respectively, parameter b _j ^~after representing secondary fuzzy assessment, whole road section traffic volume state is under the jurisdiction of the degree that corresponding j identifies traffic behavior, works as b _j ^~in while only having a value to be greater than λ, be normalized to corresponding traffic behavior; Work as b ₁ ^~, b ₂ ^~value while being all greater than λ, be normalized to " unimpeded/crowded " critical conditions; Work as b ₂ ^~, b ₃ ^~value while being all greater than λ, be normalized to " crowded/to stop up " critical conditions.

Images