CN104658252A - Method for evaluating traffic operational conditions of highway based on multisource data fusion - Google Patents

Abstract

The invention discloses a method for evaluating traffic operational conditions of a highway based on multisource data fusion and relates to the technical field of evaluation of the traffic operational conditions. The traffic operational conditions of the highway based on multisource data fusion are evaluated in combination with a large amount of GPS (global positioning system) data acquired and accumulated by a floating car acquisition system and a large amount of toll data acquired and accumulated by a highway toll system by utilizing data of fixed car detectors, and an obtained evaluation result of the traffic operational condition of the highway is more accurate and wider in coverage range, so that a defect of insufficient arrangement of car detectors can be made up for, road network monitoring range and system can be performed, traffic jams on roads can be accurately discovered in time, and the safety and efficiency of road traffic can be guaranteed; moreover, the toll data and floating car data are all based on existing toll system and floating car GPS management system, and the additional data acquisition cost is zero, so that the method provided by the embodiment of the invention is good in economy and promotion prospect.

Description

Based on the appraisal procedure of the freeway traffic running status of multisource data fusion

Technical field

The present invention relates to traffic circulation state estimation technical field, particularly relate to a kind of appraisal procedure of the freeway traffic running status based on multisource data fusion.

Background technology

The safety and efficiency of assessment to road traffic of traffic circulation state has material impact.Monitoring and prediction is carried out to traffic circulation state, in time, the traffic congestion that road exists is found accurately, focus and the difficult point of traffic circulation state estimation area research always, since the sixties in last century, one of focus of dynamic transport management research field is become to the assessment of freeway traffic running status, and has emerged in large numbers a large amount of achievements.

At present, to the research of the assessment of traffic circulation state, depend on fixed wagon detector data more, but due to the restriction by cost, the laying quantity of vehicle checker is very limited, the vehicle checker data then obtained are very limited, and traffic data is the basis of carrying out traffic circulation status monitoring and prediction, so, in prior art, limited vehicle checker data are utilized to assess traffic circulation state, the economy of traffic behavior supervision and forecast and spatial coverage is made to receive serious impact, thus cannot be timely, find the traffic congestion that road exists exactly, affect the safety and efficiency of road traffic.

Summary of the invention

The object of the present invention is to provide a kind of appraisal procedure of the freeway traffic running status based on multisource data fusion, thus solve the foregoing problems existed in prior art.

To achieve these goals, the technical solution used in the present invention is as follows:

Based on an appraisal procedure for the freeway traffic running status of multisource data fusion, comprise the steps:

Step 1, obtains the zone-to-zone travel operational factor based on multi-source data;

Step 2, merges the described zone-to-zone travel operational factor based on multi-source data, obtains the zone-to-zone travel operational factor merged;

Step 3, the zone-to-zone travel operational factor of the fusion utilizing zone-to-zone travel running status index system and step 2 to obtain, assesses zone-to-zone travel running status, obtains the assessment result of zone-to-zone travel running status;

Step 4, utilizes the place traffic circulation parameter of section traffic circulation state index system and fixing vehicle checker, assesses, obtain the assessment result of section traffic circulation state to the section traffic circulation state at fixing vehicle checker place;

Step 5, the assessment result of the section traffic circulation state that the assessment result of the zone-to-zone travel running status utilizing step 3 to obtain and step 4 obtain, obtains the assessment result of freeway traffic running status.

Preferably, in step 1, described multi-source data comprises fixing vehicle checker data, charge data and floating car data.

Preferably, in step 1, based on described fixing vehicle checker data vehicle checker section between zone-to-zone travel operational factor, utilize Cell Transmission Model to obtain.

Preferably, in step 1, based on described charge data toll station between zone-to-zone travel operational factor, by described charge data is carried out statistics obtain.

More preferably, need the described charge data of statistics to comprise: outlet charge station numbering, exit lane numbering, Outlet time, entrance charge station number, entrance lane number, entry time, vehicle, passenger-cargo classification, action type of charging and/or distance travelled; Described described charge data to be added up, be specially: to having identical outlet charge station numbering in identical Outlet time section, the vehicle of entrance charge station numbering adds up, and obtains the local train flow between toll station; The section operation speed of single car is obtained by following formula:

Section operation speed=distance travelled/(Outlet time-entry time)

Preferably, in step 1, based on the section zone-to-zone travel operational factor of described floating car data, the map-matching method of point-to-point is utilized to obtain.

Particularly, the map-matching method of described point-to-point is implemented in accordance with the following steps:

Search the driving path of Floating Car;

The physical location of Floating Car is determined in described driving path;

The data of the map-matching method input of described point-to-point comprise GPS location point data and GIS path space data.

Preferably, in step 2, by BP neural network model, the described zone-to-zone travel operational factor based on multi-source data is merged.

More preferably, described BP neural network model is made up of input layer, output layer and some hidden layers;

The input data of described input layer comprise: based on zone-to-zone travel operational factor and the sample size of various data source;

The output data of described output layer are the zone-to-zone travel operational factor after merging.

Preferably, in step 5, following formula is adopted to obtain the assessment result of freeway traffic running status:

$\hat{y}\left(t\right)=\underset{i=1}{\overset{n}{\mathrm{\Σ}}}{w}_i\left(t\right)\·{\hat{y}}_i\left(t\right)$

In formula: the assessment result of-highway running status t;

-the i-th kind of algorithm is in the assessed value of t;

W _i(t)- weight.

The invention has the beneficial effects as follows: the technical scheme that the embodiment of the present invention provides, utilize fixing vehicle checker data, gather in conjunction with Floating Car acquisition system and a large amount of gps datas accumulated, with highway tolling system collection and accumulation a large amount of charge datas, freeway traffic running status based on multisource data fusion is assessed, the assessment result of the freeway traffic running status obtained is more accurate, coverage rate is wider, vehicle checker can be made up and lay not enough defect, improve road network monitoring range and system, can realize in time, find the traffic congestion that road exists exactly, ensure the safety and efficiency of road traffic, and charge data, floating car data are all based on existing Fare Collection System, Floating Car GPS management system, and the extra acquisition cost of data is zero, so economy and the promotion prospect of the method using the embodiment of the present invention to provide are good.

Accompanying drawing explanation

Fig. 1 is the appraisal procedure schematic flow sheet that the embodiment of the present invention provides.

Embodiment

In order to make object of the present invention, technical scheme and advantage clearly understand, below in conjunction with accompanying drawing, the present invention is further elaborated.Should be appreciated that embodiment described herein only in order to explain the present invention, be not intended to limit the present invention.

As shown in Figure 1, embodiments provide a kind of appraisal procedure of the freeway traffic running status based on multisource data fusion, comprise the steps:

Step 1, obtains the zone-to-zone travel operational factor based on multi-source data;

Step 2, merges the described zone-to-zone travel operational factor based on multi-source data, obtains the zone-to-zone travel operational factor merged;

Step 3, the zone-to-zone travel operational factor of the fusion utilizing zone-to-zone travel running status index system and step 2 to obtain, assesses zone-to-zone travel running status, obtains the assessment result of zone-to-zone travel running status;

Step 4, utilizes the place traffic circulation parameter of section traffic circulation state index system and fixing vehicle checker, assesses, obtain the assessment result of section traffic circulation state to the section traffic circulation state at fixing vehicle checker place;

Step 5, the assessment result of the section traffic circulation state that the assessment result of the zone-to-zone travel running status utilizing step 3 to obtain and step 4 obtain, obtains the assessment result of freeway traffic running status.

Carry out compared with the appraisal procedure of traffic circulation state with the single vehicle checker data that utilize of the prior art, in the present embodiment, adopt and traffic circulation state estimation is carried out to the fusion of the traffic circulation parameter based on multi-source data, the assessment result of the freeway traffic running status obtained is more accurate, coverage rate is wider, vehicle checker can be made up and lay not enough defect, improve road network monitoring range and system, can realize in time, find exactly traffic congestion that road exists to ensure the safety and efficiency of road traffic.

Wherein, in step 1, described multi-source data can comprise fixing vehicle checker data, charge data and floating car data.

Due to the Fare Collection System that charge data is based on existing highway, floating car data is based on existing Floating Car GPS management system, so the appraisal procedure utilizing the embodiment of the present invention to provide, when traffic circulation state is assessed, without the need to additionally setting up the acquisition system of charge data and floating car data, also without the need to additionally gathering charge data and floating car data, data relevant to assessment traffic circulation state in charge data are only needed to extract from Fare Collection System, data relevant to assessment traffic circulation state in floating car data are extracted from GPS management system, just can directly use, so, carry out compared with the appraisal procedure of traffic circulation state with utilizing vehicle checker data in prior art, the method that the embodiment of the present invention provides, its data acquisition cost is zero, , good economy performance, there is good popularizing application prospect.

In the embodiment of the present invention, in step 1, based on described fixing vehicle checker data vehicle checker section between zone-to-zone travel operational factor, utilize Cell Transmission Model to obtain.Specifically can adopt with the following method:

Section a by time discretization, and is divided into λ by CTM _aindividual equidistant cellular, wherein the length of each cellular equals the distance that vehicle under free stream condition travels in a time step δ.Wherein, cellular 1 to cellular λ _a-1 is Hun Hang district cellular, vehicle mixed row in this region of different destination; Cellular λ _afor canalization district cellular, be divided into different queue area according to downstream road section direction, vehicle, according to the object section, downstream travelled, enters different tracks and queues up.Traffic flow temporal-spatial evolution equation in CTM can be obtained:

$y_i^a\left(k\right)=q_i^a\left(k\right)\mathrm{\δ}=\min \{{\mathrm{v\ρ}}_{i-1}^a\left(k\right)\mathrm{\δ},q_{i,\max }^a\left(k\right)\mathrm{\δ},\mathrm{\ω}({\mathrm{\ρ}}_{\mathrm{jam}}^a-{\mathrm{\ρ}}_i^a\left(k\right))\mathrm{\δ}\}---\left(1\right)$

Due to n _i(k)=ρ _ik () ν δ, then have

$y_i^a\left(k\right)=\min \{n_{i-1}^a\left(k\right),Q_i^a\left(k\right),\mathrm{\ω}(N_i^a\left(k\right)-n_i^a\left(k\right))/v\}---\left(2\right)$

Simultaneously due to flow conservation, then have

$n_i^a(k+1)=n_i^a\left(k\right)+y_i^a\left(k\right)-y_{i+1}^a\left(k\right)---\left(3\right)$

Wherein for cellular i on a of section is in the influx of period k, for the cellular i of section a is at the rate of inflow of period k, for the cellular i of section a is at the vehicle density of period k, for the cellular i of section a is in the maximum flow rate of period k, for the cellular i of section a is in the maximum flow of period k, for the cellular i of section a is at the maximum load-carrying capacity of period k, for the cellular i of section a is at the vehicle number of period k.

Calculate seemingly the block up method incured loss through delay of the method for the average instantaneous velocity in section and the average instantaneous velocity of network and compute classes similar.At period k, for any cellular i on any section a, suppose that vehicle that this cellular all flows out is all with maximal rate ν ^aflow out section, and remaining vehicle is trapped in cellular, this part car speed is 0.So the method for road-section average instantaneous velocity is calculated as follows:

${\stackrel{\‾}{v}}_a\left(v\right)=\frac{\underset{i}{\mathrm{\Σ}}{v}^a{y}_{i+1}^a\left(k\right)}{\underset{i}{\mathrm{\Σ}}{n}_i^a\left(k\right)}---\left(4\right)$

In simulation process, the trip information such as time, position of all vehicles all goes on record.Therefore, according to the time entering and leave section, average, real-time travel time and corresponding average, the real-time travel time of each paths in each section in the same time can be extrapolated not.Wherein, road-section average travel time τ _ak () can obtain by the following method:

${\mathrm{\τ}}_a\left(k\right)=\underset{\mathrm{rs}}{\mathrm{\Σ}}\underset{z\∈M^{\mathrm{rs}}}{\mathrm{\Σ}}{\mathrm{\τ}}_z^a{\mathrm{\ξ}}_z^a\left(k\right)/\underset{\mathrm{rs}}{\mathrm{\Σ}}\underset{z\∈M^{\mathrm{rs}}}{\mathrm{\Σ}}{\mathrm{\ξ}}_z^a\left(k\right)---\left(5\right)$

Wherein, M ^rsit has been the set of putting all traveler compositions between r, s so far; the actual travel time of traveler z by section a; describing traveler z and whether within the k period, enter section a, is then ${\mathrm{\ξ}}_z^a\left(k\right)=1;$ Otherwise ${\mathrm{\ξ}}_z^a\left(k\right)=0.$

When $\underset{\mathrm{rs}}{\mathrm{\Σ}}\underset{z\∈M^{\mathrm{rs}}}{\mathrm{\Σ}}{\mathrm{\ξ}}_z^a\left(k\right)=0$ Time,

${\mathrm{\τ}}_a\left(k\right)=\max \{{\widetilde{\mathrm{\τ}}}_a,{\mathrm{\τ}}_a(k-1)-1\}---\left(6\right)$

Wherein, freely the flow away line time of vehicle by section a.(4-25) Section 1 of formula illustrates that vehicle can not be less than the travel time under free stream condition by the travel time of section a, namely section 2 is by FIFO condition $(k-1)+{\mathrm{\τ}}_a(k-1)\≤k+{\mathrm{\τ}}_a\left(k\right)\⇔{\mathrm{\τ}}_a\left(k\right)\≥{\mathrm{\τ}}_a(k-1)-1$ Release.

Traveler enters road network and by path p={a in the k moment ₁, a ₂..., a _ntrip, then its Actual path travel time can calculate in the following manner:

$c_p^{\mathrm{rs}}\left(k\right)={\mathrm{\τ}}_{a_1}\left(k\right)+{\mathrm{\τ}}_{a_2}(k+{\mathrm{\τ}}_{a_1}\left(k\right))+...+{\mathrm{\τ}}_{a_n}(k+{\mathrm{\τ}}_{a_1}\left(k\right)+...+{\mathrm{\τ}}_{a_{n-1}}(k+{\mathrm{\τ}}_{a_1}\left(k\right)+...))---\left(7\right)$

In actual emulation process, the average travel time information in section needs just can obtain after emulation terminates.For the ITS application under real-time traffic conditions, can only be obtained by the Traveler Information rolling section away from, then the real-time travel time τ ' in section _ak () can calculate as follows:

${\mathrm{\τ}}_a^{\′}\left(k\right)=\underset{\mathrm{rs}}{\mathrm{\Σ}}\underset{z\∈M^{\mathrm{rs}}}{\mathrm{\Σ}}{\mathrm{\τ}}_z^a{\mathrm{\ξ}}_z^a\left(k\right)/\underset{\mathrm{rs}}{\mathrm{\Σ}}\underset{z\∈M^{\mathrm{rs}}}{\mathrm{\Σ}}{\mathrm{\ξ}}_z^a\left(k\right)---\left(8\right)$

Wherein describing traveler z and whether within the k period, leave section a, is then otherwise ${\mathrm{\ζ}}_z^a\left(k\right)=0.$

When $\underset{\mathrm{rs}}{\mathrm{\Σ}}\underset{z\∈M^{\mathrm{rs}}}{\mathrm{\Σ}}{\mathrm{\ζ}}_z^a\left(k\right)=0$ Time,

${\mathrm{\τ}}_a^{\′}\left(k\right)=\max \{{\widetilde{\mathrm{\τ}}}_a,{\mathrm{\τ}}_a^{\′}(k-1)+1\}---\left(9\right)$

Therefore k moment path p={a ₁, a ₂..., a _nreal-time route travel time can calculate in the following manner:

The instantaneous traffic flow modes in a certain moment can not embody the real driving condition in section and traffic.Therefore, road-section average traveling speed is adopted to weigh the overall traffic in section.Provide the average travel time of section a below and average travel speed computing method:

${\stackrel{\‾}{\mathrm{\τ}}}_a=\frac{\underset{\mathrm{rs}}{\mathrm{\Σ}}\underset{z\∈M^{\mathrm{rs}}}{\mathrm{\Σ}}\underset{k\∈K}{\mathrm{\Σ}}{\mathrm{\τ}}_z^a{\mathrm{\ξ}}_z^a\left(k\right)}{\underset{\mathrm{rs}}{\mathrm{\Σ}}\underset{z\∈M^{\mathrm{rs}}}{\mathrm{\Σ}}\underset{k\∈K}{\mathrm{\Σ}}{\mathrm{\ξ}}_z^a\left(k\right)}---\left(11\right)$

${\stackrel{\‾}{u}}_a=\frac{\underset{\mathrm{rs}}{\mathrm{\Σ}}\underset{z\∈M^{\mathrm{rs}}}{\mathrm{\Σ}}\underset{k\∈K}{\mathrm{\Σ}}{L_a\mathrm{\ξ}}_z^a\left(k\right)}{\underset{\mathrm{rs}}{\mathrm{\Σ}}\underset{z\∈M^{\mathrm{rs}}}{\mathrm{\Σ}}\underset{k\∈K}{\mathrm{\Σ}}{{\mathrm{\τ}}_z^a\mathrm{\ξ}}_z^a\left(k\right)}---\left(12\right)$

Wherein, L _arepresent the length of section a.

In the embodiment of the present invention, in step 1, based on described charge data toll station between zone-to-zone travel operational factor, by described charge data is carried out statistics obtain.

Need the described charge data of statistics to comprise: outlet charge station numbering, exit lane numbering, Outlet time, entrance charge station number, entrance lane number, entry time, vehicle, passenger-cargo classification, action type of charging and/or distance travelled; Described described charge data to be added up, be specially: to having identical outlet charge station numbering in identical Outlet time section, the vehicle of entrance charge station numbering adds up, and obtains the local train flow between toll station; The section operation speed of single car is obtained by following formula:

Section operation speed=distance travelled/(Outlet time-entry time) (13)

G＝[M _0.25-1.5R,M _0.25+1.5R] (14)

R＝M _0.75-M _0.25(15)

In formula (14), (15), G represents effective data intervals, and the data outside every G of dropping on all need to filter; M _0.75, M _0.25be respectively and all journey times arranged by order from small to large and is divided into the quartern, be in the value of the first, the 3rd cut-point position; R represents quartile extreme difference.Utilize formula (14), (15) set to section operation speed cleans, remove irrational data.

In the embodiment of the present invention, in step 1, based on the section zone-to-zone travel operational factor of described floating car data, the map-matching method of point-to-point is utilized to obtain.

Wherein, the matching process of point-to-point refers to and anchor point is matched the node nearest with map point geometric distance electronically or the process of shape point.

In the embodiment of the present invention, implement in accordance with the following steps, realize the map match to object Floating Car in certain time window:

Search the driving path of Floating Car;

The physical location of Floating Car is determined in described driving path.

Wherein, the fundamental purpose of searching driving path is the concrete section determining that this Floating Car is passed through in this time window, the step determining physical location by by GPS location spot projection on corresponding floating vehicle travelling path, determine the concrete locus residing for particular moment Floating Car.

In the embodiment of the present invention, the data of the map-matching method input of described point-to-point comprise GPS location point data and GIS path space data.

In the embodiment of the present invention, the concrete grammar of floating vehicle travelling transitional search can be: assuming that m GPS location point data of n-th Floating Car is in time window i wherein, x, y are the latitude and longitude coordinate of location point respectively, m=1,2 ..., M, M represent n-th Floating Car all GPS location point data total number in time interval i.Because calculating in subsequent path at least needs two location points could confirm the driving trace of a Floating Car, if M=1, represent that this Floating Car n only uploads a positional information in i-th time window, then the detection data of this Floating Car are processed as invalid data, and the map match of (n+1)th Floating Car is directly carried out in the process that algorithm skips this Floating Car.

After confirming that Floating Car n has the GPS location point data being no less than in current time window, this Floating Car is carried out to the selection of initial candidate section collection.Due to the impact of the factors such as signal, there is certain error in the position data of Floating Car, and its location point is usually around actual position in a certain region, and this region is called as error band.The system of selection of initial candidate section collection utilizes GPS error region to carry out frame choosing to road, is initial candidate section by the section that error band covers.

Error band is assumed that specific shape usually, as circle.For circle, allly drop on GPS location point circular error regions in k bar section constitute this location point initial candidate section set k=1,2 ..., K, K represent the section sum comprised in the Candidate Set of initial section.If K=0, and error range does not cover any section, then mean this location point of this Floating Car not on target road network or this location point be abnormal data, these data are removed filtration.

The transport information such as travel speed are carried out collection meter according to the time window of 5min, with above-mentioned time window for periodic duty map match, often run and once just all GPS location Point matching in this widow time are completed, carry out processing with identification floating vehicle travelling path according to the GPS position information of Floating Car simultaneously and finally obtain the particular location of Floating Car on section and corresponding traffic parameter information, start again to process next Floating Car after processing all location point information of a Floating Car in this time window, the driving trace of corresponding Floating Car in a time window on not considering when carrying out Floating Car Path Recognition in this time window.

Under normal circumstances, on road, vehicle section mean speed is similar to Normal Distribution, is designated as N (v _a, σ ²).According to the sampling theorem in mathematical statistics, the section mean speed of n table flotation motor-car normal Distribution N (v _a, σ ²/ n), and:

If the section mean speed of n table flotation motor-car with actual section mean speed v _athe error probability that is less than limits of error ε be not less than 1-α, namely

$P\left(\right|{\stackrel{\&OverBar;}{v}}_f-v_a|<{\stackrel{\&OverBar;}{v}}_f-v_a)=\mathrm{\&phi;}\left(\frac{\mathrm{\&epsiv;}\sqrt{n}}{\mathrm{\&sigma;}}\right)-\mathrm{\&phi;}(-\frac{\mathrm{\&epsiv;}\sqrt{n}}{\mathrm{\&sigma;}})\&GreaterEqual;1-\mathrm{\&alpha;}---\left(16\right)$

Can obtain

$n\&GreaterEqual;{\left[\frac{{\mathrm{\&phi;}}^{-1}(1-\frac{\mathrm{\&alpha;}}{2})\&CenterDot;\mathrm{\&sigma;}}{\mathrm{\&epsiv;}}\right]}^2={\left(\frac{Z_{\mathrm{\&alpha;}/2}\&CenterDot;\mathrm{\&sigma;}}{\mathrm{\&epsiv;}}\right)}^2---\left(17\right)$

In formula, φ (x) is Standard Normal Distribution, φ ^-1x inverse function that () is φ (x).

As can be seen from the above equation, Floating Car quantity reaches minimum samples time, result of calculation is just more accurate.But in a practical situation, in an interval computing time (as 5min), the Floating Car quantity on arbitrary section can not all reach minimum samples requirement.For this problem, the embodiment of the present invention proposes to adopt adaptive weight exponentially smooth method to carry out computation interval average velocity, and its mathematical model is

$\stackrel{\&OverBar;}{v}\left(k\right)=f\left(k\right)\&CenterDot;\stackrel{\&OverBar;}{v}(k-1)+(1-f\left(k\right))\&CenterDot;\frac{1}{n}\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}{\stackrel{\&OverBar;}{v}}_i---\left(18\right)$

$f\left(k\right)=\left\{\begin{array}{cc}1-n/n_{\min }& (0\&le;n<n_{\min })\\ 0& (n\&GreaterEqual;n_{\min })\end{array}\right.---\left(19\right)$

In formula, for the section mean speed estimated value at interval current computing time; for the section mean speed estimated value at interval last computing time; F (k) is adaptive weighting.

In the embodiment of the present invention, in step 2, by BP neural network model, the described zone-to-zone travel operational factor based on multi-source data is merged.

Wherein, described BP neural network model is made up of input layer, output layer and some hidden layers;

The input data of described input layer comprise: based on zone-to-zone travel operational factor and the sample size of various data source;

The output data of described output layer are the zone-to-zone travel operational factor after merging.

In a preferred embodiment of the invention, adopt BP neural net method, average velocity is run to the zone-to-zone travel of fixing vehicle checker data, charge data and floating car data and carries out information fusion.The input data of described input layer comprise: the section mean speed that floating car data calculates, the sample size of floating car data, section mean speed, the sample size data of charge data, the fixing section mean speed of vehicle checker data calculating and the sample size of fixing vehicle checker data that charge data calculates.Therefore, the input data of neural network are above-mentioned 6 parameters, and the output data of output layer are the section mean speed after merging.According to input data and the output data of neural network, determine that the input layer of neural network comprises 6 nodes (section mean speed that the section mean speed that floating car data calculates, Floating Car sample size, charge data calculate, the sample size data of charge data, the fixing section mean speed of vehicle checker data calculating and the sample size of fixing vehicle checker data), output layer only comprises a node (exporting the section mean speed after merging), and hidden layer nodal point number is 9.

In the embodiment of the present invention, to congested in traffic degree index norm-setting in " network of highways traffic circulation monitoring with service provisional technical requirement " that traffic circulation state index system is promulgated according to Department of Transportation.

In the embodiment of the present invention, after obtaining the zone-to-zone travel operational factor merged, implementation step 3, the zone-to-zone travel operational factor of the fusion utilizing zone-to-zone travel running status index system and step 2 to obtain, zone-to-zone travel running status is assessed, obtains the assessment result of zone-to-zone travel running status.

In the preferred embodiment of the present invention, zone-to-zone travel running status index system (express highway section crowding index) is as shown in table 1.

Table 1 express highway section crowding index (overall speed index)

In the zone-to-zone travel running status index system of section mean speed shown in table 1 that determining step 2 exports, the interval range at place, obtains the traffic circulation state in section, exports the zone-to-zone travel running statuses such as unimpeded, substantially unimpeded, general, crowded, blocking.

In the embodiment of the present invention, section traffic circulation state index system (express highway section crowding index) is as shown in table 2.

In the preferred embodiments of the present invention, implementation step 4 with the following method can be adopted.

Utilize the place traffic circulation parameter (the vehicle checker section of 5 minutes statistics divides directional flow, speed, occupation rate) of fixing vehicle checker, by the traffic circulation state of the section traffic circulation state index system assessment vehicle checker place section of table 2, interval according to the residing in table 2 speed of vehicle checker profile data, occupation rate, judge the traffic circulation state (block up, walk or drive slowly, unimpeded) of section representated by vehicle checker.

Table 2 section traffic circulation state index system (section speed-occupation rate index)

In the embodiment of the present invention, the assessment result of the section traffic circulation state that the assessment result of the zone-to-zone travel running status utilizing step 3 to obtain and step 4 obtain, obtains the assessment result of freeway traffic running status.Following formula specifically can be adopted to obtain the assessment result of freeway traffic running status:

$\hat{y}\left(t\right)=\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}{w}_i\left(t\right)\&CenterDot;{\hat{y}}_i\left(t\right)$

In formula:

the assessment result of-highway running status t;

-the i-th kind of algorithm is in the assessed value of t;

W _i(t)- weight.

In the present embodiment, in above-mentioned formula, according to the assessment result of different pieces of information source algorithm (comprising the algorithm based on floating car data, the algorithm based on charge data, algorithm based on fixing vehicle checker data) the weighted value w of the ratio calculating of total sample size is accounted in conjunction with the sample size according to this algorithm _it (), obtains final highway condition evaluation results by weighted mean

By adopting technique scheme disclosed by the invention, obtain effect useful as follows:: the technical scheme that the embodiment of the present invention provides, utilize fixing vehicle checker data, gather in conjunction with Floating Car acquisition system and a large amount of gps datas accumulated, with highway tolling system collection and accumulation a large amount of charge datas, freeway traffic running status based on multisource data fusion is assessed, the assessment result of the freeway traffic running status obtained is more accurate, coverage rate is wider, vehicle checker can be made up and lay not enough defect, improve road network monitoring range and system, can realize in time, find the traffic congestion that road exists exactly, ensure the safety and efficiency of road traffic.

Each embodiment in this instructions all adopts the mode of going forward one by one to describe, and what each embodiment stressed is the difference with other embodiments, between each embodiment identical similar part mutually see.

Those skilled in the art it should be understood that the sequential of the method step that above-described embodiment provides can carry out accommodation according to actual conditions, also can carry out according to actual conditions are concurrent.

The hardware that all or part of step in the method that above-described embodiment relates to can carry out instruction relevant by program has come, described program can be stored in the storage medium that computer equipment can read, for performing all or part of step described in the various embodiments described above method.Described computer equipment, such as: personal computer, server, the network equipment, intelligent mobile terminal, intelligent home device, wearable intelligent equipment, vehicle intelligent equipment etc.; Described storage medium, such as: the storage of RAM, ROM, magnetic disc, tape, CD, flash memory, USB flash disk, portable hard drive, storage card, memory stick, the webserver, network cloud storage etc.

Finally, also it should be noted that, in this article, the such as relational terms of first and second grades and so on is only used for an entity or operation to separate with another entity or operational zone, and not necessarily requires or imply the relation that there is any this reality between these entities or operation or sequentially.And, term " comprises ", " comprising " or its any other variant are intended to contain comprising of nonexcludability, thus make to comprise the process of a series of key element, method, commodity or equipment and not only comprise those key elements, but also comprise other key elements clearly do not listed, or also comprise by the intrinsic key element of this process, method, commodity or equipment.When not more restrictions, the key element limited by statement " comprising ... ", and be not precluded within process, method, commodity or the equipment comprising described key element and also there is other identical element.

The above is only the preferred embodiment of the present invention; it should be pointed out that for those skilled in the art, under the premise without departing from the principles of the invention; can also make some improvements and modifications, these improvements and modifications also should look protection scope of the present invention.

Claims (10)

1., based on an appraisal procedure for the freeway traffic running status of multisource data fusion, it is characterized in that, comprise the steps:

Step 1, obtains the zone-to-zone travel operational factor based on multi-source data;

Step 2, merges the described zone-to-zone travel operational factor based on multi-source data, obtains the zone-to-zone travel operational factor merged;

Step 3, the zone-to-zone travel operational factor of the fusion utilizing zone-to-zone travel running status index system and step 2 to obtain, assesses zone-to-zone travel running status, obtains the assessment result of zone-to-zone travel running status;

Step 4, utilizes the place traffic circulation parameter of section traffic circulation state index system and fixing vehicle checker, assesses, obtain the assessment result of section traffic circulation state to the section traffic circulation state at fixing vehicle checker place;

Step 5, the assessment result of the section traffic circulation state that the assessment result of the zone-to-zone travel running status utilizing step 3 to obtain and step 4 obtain, obtains the assessment result of freeway traffic running status.

2. the appraisal procedure of the freeway traffic running status based on multisource data fusion according to claim 1, is characterized in that, in step 1, described multi-source data comprises fixing vehicle checker data, charge data and floating car data.

3. the appraisal procedure of the freeway traffic running status based on multisource data fusion according to claim 2, it is characterized in that, in step 1, based on described fixing vehicle checker data vehicle checker section between zone-to-zone travel operational factor, utilize Cell Transmission Model to obtain.

4. the appraisal procedure of the freeway traffic running status based on multisource data fusion according to claim 2, it is characterized in that, in step 1, based on described charge data toll station between zone-to-zone travel operational factor, by described charge data is carried out statistics obtain.

5. the appraisal procedure of the freeway traffic running status based on multisource data fusion according to claim 4, it is characterized in that, need the described charge data of statistics to comprise: outlet charge station numbering, exit lane numbering, Outlet time, entrance charge station number, entrance lane number, entry time, vehicle, passenger-cargo classification, action type of charging and/or distance travelled; Described described charge data to be added up, be specially: to having identical outlet charge station numbering in identical Outlet time section, the vehicle of entrance charge station numbering adds up, and obtains the local train flow between toll station; The section operation speed of single car is obtained by following formula:

Section operation speed=distance travelled/(Outlet time-entry time)

6. the appraisal procedure of the freeway traffic running status based on multisource data fusion according to claim 2, it is characterized in that, in step 1, based on the section zone-to-zone travel operational factor of described floating car data, the map-matching method of point-to-point is utilized to obtain.

7. the appraisal procedure of the freeway traffic running status based on multisource data fusion according to claim 6, it is characterized in that, the map-matching method of described point-to-point is implemented in accordance with the following steps:

Search the driving path of Floating Car;

The physical location of Floating Car is determined in described driving path;

The data of the map-matching method input of described point-to-point comprise GPS location point data and GIS path space data.

8. the appraisal procedure of the freeway traffic running status based on multisource data fusion according to claim 1, is characterized in that, in step 2, is merged the described zone-to-zone travel operational factor based on multi-source data by BP neural network model.

9. the appraisal procedure of the freeway traffic running status based on multisource data fusion according to claim 8, is characterized in that, described BP neural network model is made up of input layer, output layer and some hidden layers;

The input data of described input layer comprise: based on zone-to-zone travel operational factor and the sample size of various data source;

The output data of described output layer are the zone-to-zone travel operational factor after merging.

10. the appraisal procedure of the freeway traffic running status based on multisource data fusion according to claim 1, is characterized in that, in step 5, adopts following formula to obtain the assessment result of freeway traffic running status:

$\hat{y}\left(t\right)=\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}{w}_i\left(t\right)\&CenterDot;{\hat{y}}_i\left(t\right)$

In formula: the assessment result of-highway running status t;

-the i-th kind of algorithm is in the assessed value of t;

W _i(t)- weight.