CN102737504B - Method for estimating bus arrival time in real time based on drive characteristics - Google Patents

Abstract

The invention discloses a method for estimating bus arrival time in real time based on drive characteristics. The method comprises the following steps: firstly, analyzing the drive characteristics to obtain a drive characteristic correction factor; secondly, obtaining a travel time estimation value according to the average vehicle speed, the average travel time and the correction factor; thirdly, estimating bus station delay time and signal lamp delay time; and lastly, obtaining estimated bus arrival time. According to the method, bus driving time on a road network is divided into road section travel time and delay time, accurate estimation of the road section travel time and reasonable calculation of bus stop time and the signal lamp delay time are determining factors for estimating arrival time of a bus floating vehicle, and the sum of time of two parts is the arrival time of the bus floating vehicle from the current position to the target position. Different road section data quantities are calculated by different algorithms respectively, so that the prediction accuracy of the average travel time of the road section and the prediction accuracy of the bus arrival time are improved.

Description

A kind of bus arrival time real-time estimation method based on driving performance

Technical field

The present invention relates to intelligent transportation system technical field, particularly a kind of bus arrival time real-time estimation method based on driving performance.

Background technology

In intelligent transportation system research, the factor analysis such as the section mean speed in the estimation of bus arrival time and the magnitude of traffic flow, section, current period, road conditions and distance leaving from station.Based on the bus data that can be used for bus arrival time estimation, the real-time information of these data comprises: vehicle instantaneous velocity (characterizing the speed of vehicle current time), distance travelled (characterizing the total kilometrage that vehicle travels from GPS device is installed), mileage between standing (characterize vehicle in once operation process from inception point to the distance travelled of Current GPS anchor point), direction of traffic (characterizes the vehicle direction of dispatching a car, for example: inception point is decided to be " 1 " to the direction of traffic of terminus, terminus is decided to be " 0 " to the direction of traffic of inception point), vehicle-state (characterizes the inbound case of vehicle, departures situation, arrive at a station situation and type of site etc.), longitude and latitude position, current time (characterizes the date of Current GPS anchor point, the information such as Hour Minute Second) and limiting vehicle speed value etc., can carry out preferably the estimation of bus arrival time.

Bus arrival time method of estimation in the past has static prediction and two kinds of patterns of performance prediction.Static method is difficult to adapt to road traffic state complicated and changeable.And in dynamic approach, on the one hand, the instantaneous velocity of vehicle is subject to Assessment on Environmental Impact Affected and shows larger randomness and mutability, data have considerable influence to prediction more greatly, more frequently in fluctuation, will cause the unreliability of prediction; On the other hand, bus travels in route, and road inherent condition is more stable, and in one day, road condition variation can present certain regularity, as: morning peak and evening peak, and along with weather also has certain variation, this rule can have influence on driver's driving performance, make its driving performance there is certain regularity, and for different drivers, this rule is also not quite similar, and therefore, for improving the precision of estimating, this rule also needs to consider.

Meanwhile, due to the impact of road environment and wagon flow, bus exists the delay time at stop under steam, delay time at stop mainly comprises two parts, the one, the arriving at a station the residence time of bus, particularly peak period and off-peak period exist significant difference, therefore need to consider respectively; The 2nd, the signal lamp delay time at stop.

Therefore be badly in need of the method for calculating bus time of arrival on different pieces of information sample size basis a kind of.

Summary of the invention

In view of this, technical matters to be solved by this invention is to provide the method for calculating bus time of arrival on different pieces of information sample size basis a kind of.

Implementation procedure of the present invention is as follows:

Bus arrival time real-time estimation method based on driving performance provided by the invention, adopts following formula to calculate bus arrival time:

Wherein: T _ntkfor t moment public transit vehicle k gets to the station estimated time of n from current location; T _tfor the average travel time of t moment public transit vehicle from current location to downstream targets website;

for stopping the delay time at stop of vehicle x; D _pfor vehicle is in the signal lamp place stop delay time.

Further, the average travel time T of described t moment public transit vehicle from current location to downstream targets website _tthe average travel time that comprises car day shift is estimated;

Described day shift, the average travel time of car was estimated to calculate by following formula:

$\stackrel{\‾}{T_{\mathrm{travel}}^b}={\mathrm{\α}}_{(j,x)}^k\frac{l}{\stackrel{\‾}{V_{(j,x)}}}+\underset{i=j+1}{\overset{n}{\mathrm{\Σ}}}\frac{{\mathrm{\α}}_{(j,x)}^k{L}_i}{\stackrel{\‾}{V_{(i,x)}}};$

Wherein,

for revising average travel time, for the section mean speed of the section j between vehicle and nearest website;

for the section mean speed of section i; J is section, the current place of vehicle; L _ifor the length of section i; N is a section before target station;

under situation k, to the driving performance modifying factor of the upper vehicle x of section i.

The section mean speed in described section

to calculate by following steps:

First take station as unit divides section, obtain site information and direction of traffic information in gps data, then obtain the identical all floating car datas of direction of estimating with average travel time in this section within this time period, finally ask for the section mean speed in this section by floating car data

Further, the average travel time T of described t moment public transit vehicle from current location to downstream targets website _tthe average travel time that also comprises night shift bus is estimated;

The average travel time of described night shift bus estimates to comprise that short distance Target Station average travel time is estimated and long distance objective station average travel time is estimated;

Described short distance Target Station average travel time is estimated to calculate by following formula:

$\stackrel{\‾}{T_{\mathrm{travel}}^n}=\frac{L_d}{\stackrel{\‾}{V}};$

Wherein, for this night shift bus is to the average travel time of downstream targets website, L _drepresent to utilize the vehicle mileage information in gps data to carry out the distance between definite vehicle-to-target website,

represent the average velocity in this time period;

Described long distance objective station average travel time is estimated to calculate by following formula:

$\stackrel{\‾}{T_{\mathrm{travel}}^n}=\frac{1}{\stackrel{\‾}{V}}+\underset{i=j}{\overset{n}{\mathrm{\Σ}}}{t}_{(i,r)}^k;$

Wherein, the previous section that n is targeted sites; J is the next section that vehicle will arrive;

for under situation k, vehicle is through running time between the required average station of section i, and l is the distance between vehicle and nearest website.

Described average velocity

calculate by following formula:

$\stackrel{\‾}{V}=\frac{1}{M}\underset{i=1}{\overset{M}{\mathrm{\Σ}}}{v}_i;$

Wherein, M is the GPS valid data number of this vehicle of collecting, v _ifor speed corresponding to each vehicle in gps data.

Further, stopping the delay time at stop of described vehicle x

calculate by following formula:

$\stackrel{\‾}{D_x}=D_b+\underset{i=j}{\overset{n}{\mathrm{\Σ}}}{t}_{(i,s)}^k;$

Wherein, the previous station that n is targeted sites; J is the downstream station nearest from vehicle current location;

for under situation k, vehicle is the mean residence time of i AT STATION;

Described each station is in the average dwell time of different periods

calculate by following formula:

D _b＝αT ₀+βT ₁+γT ₂；

Wherein: α, beta, gamma is time correction factor; T ₀pull in the time of sailing into for slowing down; T ₁for the boarding and alighting time that stops; T ₂for accelerating leaving from station rolling away from the time.

Further, described vehicle is in signal lamp place stop delay time D _pcalculate by following steps:

S51: the public transit vehicle position location and the instantaneous velocity that utilize gps data Real-time Obtaining;

S52: determine that whether public transit vehicle is in signal lamp region;

S53: according to the transport condition of vehicle, calculate the residence time of vehicle in current traffic lights position by following formula:

S54: in the time that vehicle is met red light, from current location to the stop delay time sum of estimating all signal lamps station be $D_p=\underset{p=1}{\overset{q}{\mathrm{\Σ}}}\underset{x=1}{\overset{M}{\mathrm{\Σ}}}P\left(x\right)T_{\mathrm{px}};$

S55: when vehicle is met when green light, start from the next signal lamp of current location to the stop delay time sum of estimating all signal lamps station be $D_p=\underset{p=2}{\overset{q}{\mathrm{\Σ}}}\underset{x=1}{\overset{M}{\mathrm{\Σ}}}P\left(x\right)T_{\mathrm{px}};$

S56: when vehicle is not during or not signal lamp region, from current location to the stop delay time sum of estimating all signal lamps station be $D_p=\underset{p=1}{\overset{q}{\mathrm{\Σ}}}\underset{x=1}{\overset{M}{\mathrm{\Σ}}}P\left(x\right)T_{\mathrm{px}};$

Wherein, in the time period that P (x) expression time interval length is t, there is the probability distribution in x car arriving signal lamp region, T _pxfor the random delay time at stop of vehicle parking at signal lamp p place, M is that the vehicle in signal lamp p place one-period arrives number.

Further, described driving performance modifying factor

obtain by following steps:

S61: obtain vehicle site information and information out of the station in gps data;

S62: average running time between the station of calculating website;

S63: according to the ratio of running time between vehicle place cluster centre and average station, draw the driving performance modifying factor of vehicle.

The invention has the advantages that: the present invention adopts bus is mainly divided into two parts at road network running time: Link Travel Time and delay time at stop, high precision estimation and the reasonable computation to bus dwell time and signal lamp delay time at stop to current time current location bus to the average travel time for road sections of downstream website, two parts time is combined is the time of arrival of public transport Floating Car from current location to target location, section for different pieces of information amount adopts respectively the instantaneous velocity method of substitution, the compound integral method of instantaneous velocity, the algorithms of different such as the compound section method of weighting are calculated the precision of prediction that improves average travel time for road sections, improve the precision of prediction of bus arrival time.High precision estimation and the reasonable computation to bus dwell time and signal lamp delay time at stop to current time current location public transport Floating Car to the average travel time for road sections of downstream website are the deciding factors that determines estimation time of arrival of public transport Floating Car.

Three aspects of main consideration in the average travel time for road sections specificity analysis of public transport Floating Car: the Link Travel Time of current period of 1. public transport Floating Car is inevitable relevant with current current section, between the unsteady station of public transport, along with the change of current road section traffic volume traffic status, difference has the different rules of travelling to journey time in the section with different geographic entitys; 2. at dead of night, the traffic traffic status facing of night shift bus and daytime, there were significant differences, the traffic traffic status in the late into the night is simpler than daytime; The Link Travel Time of 3. public transport Floating Car also can be associated with the driving performance of current driver's self, in the same period, and the section of identical geographic entity, different drivers can have different driving performances.

Two aspects of main consideration in the delay time at stop specificity analysis of public transport Floating Car: having the traffic lights of some in 1. public transport Floating Car route, is to have certain randomness in the delay in traffic lights region; 2. in public transport Floating Car route, have the bus stop of some, vehicle exists significant difference in the residence time of bus stop in different weather condition and different periods, particularly peak period and off-peak period, therefore needs to consider respectively.

Other advantage of the present invention, target and feature will be set forth to a certain extent in the following description, and to a certain extent, based on will be apparent to those skilled in the art to investigating below, or can be instructed from the practice of the present invention.The objects and other advantages of the present invention can be passed through instructions below, claims, and in accompanying drawing, specifically noted structure realizes and obtains.

Accompanying drawing explanation

In order to make the object, technical solutions and advantages of the present invention clearer, below in conjunction with accompanying drawing, the present invention is described in further detail, wherein:

The method figure that Fig. 1 estimates for the bus arrival time that the embodiment of the present invention provides.

Embodiment

Below with reference to accompanying drawing, the preferred embodiments of the present invention are described in detail; Should be appreciated that preferred embodiment is only for the present invention is described, rather than in order to limit the scope of the invention.

The method figure that Fig. 1 estimates for the bus arrival time that the embodiment of the present invention provides, as shown in the figure: the bus arrival time real-time estimation method based on driving performance provided by the invention, adopts following formula to calculate bus arrival time:

Wherein: T _ntkfor t moment public transit vehicle k gets to the station estimated time of n from current location; T _tfor the average travel time of t moment public transit vehicle from current location to downstream targets website;

for stopping the delay time at stop of vehicle x; D _pfor vehicle is in the signal lamp place stop delay time.

The average travel time T of described t moment public transit vehicle from current location to downstream targets website _tthe average travel time that comprises car day shift is estimated;

Described day shift, the average travel time of car was estimated to calculate by following formula:

$\stackrel{\‾}{T_{\mathrm{travel}}^b}={\mathrm{\α}}_{(j,x)}^k\frac{l}{\stackrel{\‾}{V_{(j,x)}}}+\underset{i=j+1}{\overset{n}{\mathrm{\Σ}}}\frac{{\mathrm{\α}}_{(j,x)}^k{L}_i}{\stackrel{\‾}{V_{(i,x)}}};$

Wherein,

for revising average travel time,

for the section mean speed of the section j between vehicle and nearest website;

for the section mean speed of section i; J is section, the current place of vehicle; L _ifor the length of section i; N is a section before target station;

under situation k, to the driving performance modifying factor of the upper vehicle x of section i.

The section mean speed in described section to calculate by following steps:

First take station as unit divides section, obtain site information and direction of traffic information in gps data, then obtain the identical all floating car datas of direction of estimating with average travel time in this section within this time period, finally ask for the section mean speed in this section by floating car data

The average travel time T of described t moment public transit vehicle from current location to downstream targets website _tthe average travel time that also comprises night shift bus is estimated;

The average travel time of described night shift bus estimates to comprise that short distance Target Station average travel time is estimated and long distance objective station average travel time is estimated;

Described short distance Target Station average travel time is estimated to calculate by following formula:

$\stackrel{\‾}{T_{\mathrm{travel}}^n}=\frac{L_d}{\stackrel{\‾}{V}};$

Wherein,

for this night shift bus is to the average travel time of downstream targets website, L _drepresent to utilize the vehicle mileage information in gps data to carry out the distance between definite vehicle-to-target website,

represent the average velocity in this time period;

Described long distance objective station average travel time is estimated to calculate by following formula:

$\stackrel{\‾}{T_{\mathrm{travel}}^n}=\frac{1}{\stackrel{\‾}{V}}+\underset{i=j}{\overset{n}{\mathrm{\Σ}}}{t}_{(i,r)}^k;$

Wherein, the previous section that n is targeted sites; J is the next section that vehicle will arrive;

for under situation k, vehicle is through running time between the required average station of section i, and l is the distance between vehicle and nearest website.

Described average velocity

calculate by following formula:

$\stackrel{\‾}{V}=\frac{1}{M}\underset{i=1}{\overset{M}{\mathrm{\Σ}}}{v}_i;$

Wherein, M is the GPS valid data number of this vehicle of collecting, v _ifor speed corresponding to each vehicle in gps data.

Described vehicle x stops the delay time at stop

calculate by following formula:

$\stackrel{\‾}{D_x}=D_b+\underset{i=j}{\overset{n}{\mathrm{\Σ}}}{t}_{(i,s)}^k;$

Wherein, the previous station that n is targeted sites; J is the downstream station nearest from vehicle current location;

for under situation k, vehicle is the mean residence time of i AT STATION;

The described vehicle delay time at stop D that stops in the region, station of current station b _bcalculate by following formula:

D _b＝αT ₀+βT ₁+γT ₂；

Wherein: α, beta, gamma is time correction factor; T ₀pull in the time of sailing into for slowing down; T ₁for the boarding and alighting time that stops; T ₂for accelerating leaving from station rolling away from the time.

Vehicles while passing station time, i.e. T ₀and T ₂available empirical value replaces.?

wherein,

for under situation k, vehicle is the averaging time of the interior stop of b AT STATION.

In the time that vehicle deceleration enters the station, now α ∈ (0,1) (representing that α gets the random value between 0～1), β=γ=1;

When vehicle is when region stops AT STATION, now α=0, β ∈ (0,1) (represent β get the random value between 0～1), γ=1;

When vehicle accelerates time leaving from station, now α=0, β=0, γ ∈ (0,1) (representing that γ gets the random value between 0～1);

When vehicle is not AT STATION when region, now α=β=γ=0.

Described each station is in the average dwell time of different periods

according to the information out of the station of vehicle in new terminal gps data, then utilize these a large amount of historical datas, the data out of the station of calculating vehicle, obtain its temporal information, deduct the time of entering the station obtain the dwell time of this car at this website with the departures time of same website.Then obtain respectively in fine day peak period, fine day off-peak period, peak period rainy day and off-peak period rainy day these 4 kinds of situations the average dwell time of each website

(wherein k represents above-mentioned 4 kinds of situations, and i represents car station number).Between its analytical approach and vehicle station, the analytical approach of average running time is identical, therefore repeats no more.

Described vehicle is in signal lamp place stop delay time D _pcalculate by following steps:

Suppose that in the signal period, red time is T _h, T _hcan be obtained by reality investigation.T _pxbe defined as the random delay time at stop of vehicle parking at signal lamp p place, known T _pxa random number and T _px∈ (0, T _h).

Suppose the arrival obedience Poisson distribution of vehicle in section, therefore have:

$P\left(x\right)=\frac{{\left(\mathrm{\λt}\right)}^x{e}^{-\mathrm{\λt}}}{x!}=\frac{d^x{e}^{-d}}{x!},t>0,x=0,\mathrm{1,2},...M;$

Above formula represents that time interval length is the probability that has x car arriving signal lamp region in time period of t; The vehicle mean arrival rate at λ representation unit interval; D represents vehicle mean arrival rate in the t of gate time interval.The vehicle that M is defined as in signal lamp p place one-period arrives number.

S51: the public transit vehicle position location and the instantaneous velocity that utilize gps data Real-time Obtaining;

S52: determine that whether public transit vehicle is in signal lamp region;

S53: according to the transport condition of vehicle, calculate the residence time of vehicle in current traffic lights position by following formula:

S54: in the time that vehicle is met red light, from current location to the stop delay time sum of estimating all signal lamps station be $D_p=\underset{p=1}{\overset{q}{\mathrm{\Σ}}}\underset{x=1}{\overset{M}{\mathrm{\Σ}}}P\left(x\right)T_{\mathrm{px}};$

S55: when vehicle is met when green light, start from the next signal lamp of current location to the stop delay time sum of estimating all signal lamps station be $D_p=\underset{p=2}{\overset{q}{\mathrm{\Σ}}}\underset{x=1}{\overset{M}{\mathrm{\Σ}}}P\left(x\right)T_{\mathrm{px}};$

S56: when vehicle is not during or not signal lamp region, from current location to the stop delay time sum of estimating all signal lamps station be $D_p=\underset{p=1}{\overset{q}{\mathrm{\Σ}}}\underset{x=1}{\overset{M}{\mathrm{\Σ}}}P\left(x\right)T_{\mathrm{px}};$

Wherein, in the time period that P (x) expression time interval length is t, there is the probability distribution in x car arriving signal lamp region, T _pxfor the random delay time at stop of vehicle parking at signal lamp p place, M is that the vehicle in signal lamp p place one-period arrives number.

Described driving performance modifying factor

obtain by following steps:

S61: obtain vehicle site information and information out of the station in gps data; Utilize the data out of the station of gps data between this adjacent sites, the departures time that deducts previous station by the time of entering the station at a rear station obtain running time between the station of this adjacent sites and obtain its temporal information.

S62: average running time between the station of calculating website; Between the station of website, running time comprises between bicycle station average running time between average running time and many stations; By running time data between the sorted station that between station, running time raw data obtains after by hierarchical clustering, thereby obtain the weight of running time between each interval station, utilize this weight each district section vehicle running time between the average station in this section that superposes,

Following example 1 is the calculating embodiment of average running time between bicycle station:

Example 1: establish off-peak period, vehicle x is running time data (unit: second) between the station of one group of fine day of section i, and as shown in table 1 below is running time raw data between station, calculates average running time between this station, and statistics is as shown in table 1:

Table 1

As shown in table 2 by running time data between the sorted station obtaining after hierarchical clustering:

Table 2

The weight that obtains running time between each interval station is:

$f_1=f_2=\frac{3^2}{3^2+3^2+5^2+2^2}=\frac{9}{47};$

$f_3=\frac{5^2}{3^2+3^2+5^2+2^2}=\frac{25}{47};$

$f_4=\frac{2^2}{3^2+3^2+5^2+2^2}=\frac{4}{47};$

This car running time between the average station in this section is:

$t_{(i,x,r)}^k=\frac{9}{47}\×\left(\frac{97+112+116}{3}\right)+\frac{9}{47}\×\left(\frac{120+133+139}{3}\right)$

$+\frac{25}{47}\×\left(\frac{140+145+147+149+149}{5}\right)+\frac{4}{47}\×\left(\frac{190+251}{2}\right)$

$=142.2$

Following example 2 is the calculating embodiment of average running time between many stations:

Can obtain between the station of different vehicle on the i of section average running time by method shown in example 1, analyze by certain method more on this basis, ask for running time between the average station of section i.Method is as shown in example 2.

Example 2: be located at different vehicle that fine day off-peak period statistics the obtains section i running time (unit: second) of on average standing, as shown in the table, calculate running time between the average station of section i.Statistics is as shown in table 3:

Running time statistics between table 3 station

The classification results that the data in section 1 obtain after to vehicle classification by hierarchical clustering is as shown in table 4, in bracket, is corresponding car number:

Result after table 4 classification

Can be drawn by method shown in example 1, between the average station of section i, running time is

second.

Finally respectively for fine day peak period, fine day off-peak period, peak period rainy day and off-peak period rainy day these 4 kinds of situations, obtain respectively average running time between the station in each section of these 4 kinds of situations with said method

(wherein k represents above-mentioned 4 kinds of situations, and i represents road segment number).

S63: on average the stand class bunch of a running time of the vehicle obtaining according to hierarchical clustering, adopt mean value method to calculate each Lei Cu center, and using between the average station of this value all vehicles in such bunch running time.Then according to the ratio of running time between the average station in running time and section between the average station of vehicle, draw the driving performance modifying factor of vehicle.

According to the statistics of average running time between the station of vehicle, analyze, show that respectively vehicle is in above-mentioned 4 kinds of situations, i.e. fine day peak period, fine day off-peak period, peak period rainy day and off-peak period rainy day, the modifying factor of the driving performance on different sections of highway.Analytical approach is as shown in example 3.

Example 3: establish the vehicle of adding up the fine day peak absences obtaining by the historical data running time (unit: second) of on average standing, as shown in the table, calculate the driving performance modifying factor of vehicle 1 in each section

(wherein, k represents above-mentioned 4 kinds of situations, and i represents section numbering, and x represents car number).Statistics is as shown in table 5:

Average running time statistics between table 5 station

Can be drawn by method shown in example 2, between the average station in section 1, running time is 261.4 seconds.The now center C of class bunch 1 ₁equal the mean value of data in class bunch 1,

second.By the ratio of running time between vehicle 1 place cluster centre and average station, can show that a modifying factor component of vehicle 1 is

in like manner can obtain 8 sections calculating below respectively, ${\mathrm{\α}}_{\left(\mathrm{2,1}\right)}^2=0.641,$ ${\mathrm{\α}}_{\left(\mathrm{3,1}\right)}^2=0.839,$ ${\mathrm{\α}}_{\left(\mathrm{4,1}\right)}^2=0.907,$ ${\mathrm{\α}}_{\left(\mathrm{5,1}\right)}^2=1.013,$ ${\mathrm{\α}}_{\left(\mathrm{6,1}\right)}^2=0.664,$ ${\mathrm{\α}}_{\left(\mathrm{7,1}\right)}^2=1.087,$ ${\mathrm{\α}}_{\left(\mathrm{8,1}\right)}^2=0.908,$ ${\mathrm{\α}}_{\left(\mathrm{9,1}\right)}^2=0.875.$

Therefore, the driving performance modifying factor value of vehicle 1, as shown in table 6.

The modifying factor of table 6 vehicle 1 in the situation of the non-peak of fine day

The foregoing is only the preferred embodiments of the present invention, be not limited to the present invention, obviously, those skilled in the art can carry out various changes and modification and not depart from the spirit and scope of the present invention the present invention.Like this, if within of the present invention these are revised and modification belongs to the scope of the claims in the present invention and equivalent technologies thereof, the present invention is also intended to comprise these changes and modification interior.

Claims (4)

1. the bus arrival time real-time estimation method based on driving performance, is characterized in that: adopt following formula to calculate bus arrival time:

Wherein: T _ntkfor t moment public transit vehicle k gets to the station estimated time of n from current location; T _tfor the average travel time of t moment public transit vehicle from current location to downstream targets website;

for stopping the delay time at stop of vehicle x; D _pfor vehicle is in the signal lamp place stop delay time;

The average travel time T of described t moment public transit vehicle from current location to downstream targets website _tthe average travel time that comprises car day shift is estimated;

Described day shift, the average travel time of car was estimated to calculate by following formula:

$\stackrel{\‾}{T_{\mathrm{travel}}^b}={\mathrm{\α}}_{(j,x)}^k\frac{l}{\stackrel{\‾}{V_{(j,x)}}}+\underset{i=j+1}{\overset{n}{\mathrm{\Σ}}}\frac{{\mathrm{\α}}_{(j,x)}^k{L}_i}{\stackrel{\‾}{V_{(i,x)}}};$

Wherein,

for revising average travel time,

for the section mean speed of the section j between vehicle and nearest website;

for the section mean speed of section i; J is section, the current place of vehicle; L _ifor the length of section i; N is a section before target station;

under situation k, to the driving performance modifying factor of the upper vehicle x of section i;

The section mean speed in described section

to calculate by following steps:

First take station as unit divides section, obtain site information and direction of traffic information in gps data, then obtain the identical all floating car datas of direction of estimating with average travel time in this section within this time period, finally ask for the section mean speed in this section by floating car data

Described driving performance modifying factor

obtain by following steps:

S61: obtain vehicle site information and information out of the station in gps data;

S62: average running time between the station of calculating website;

S63: by the ratio of average running time between the station of Lei Cu center, vehicle place and website after the chronological classification of on average travelling between station, be the driving performance modifying factor of vehicle.

2. the bus arrival time real-time estimation method based on driving performance according to claim 1, is characterized in that: the average travel time T of described t moment public transit vehicle from current location to downstream targets website _tthe average travel time that also comprises night shift bus is estimated;

The average travel time of described night shift bus estimates to comprise that short distance Target Station average travel time is estimated and long distance objective station average travel time is estimated;

Described short distance Target Station average travel time is estimated to calculate by following formula:

$\stackrel{\‾}{T_{\mathrm{travel}}^n}=\frac{L_d}{\stackrel{\‾}{V}};$

Wherein,

for this night shift bus is to the average travel time of downstream targets website, L _drepresent to utilize the vehicle mileage information in gps data to carry out the distance between definite vehicle-to-target website,

represent the average velocity in this time period;

Described long distance objective station average travel time is estimated to calculate by following formula:

$\stackrel{\‾}{T_{\mathrm{travel}}^n}=\frac{l}{\stackrel{\‾}{V}}+\underset{i=j}{\overset{n}{\mathrm{\Σ}}}{t}_{(i,r)}^k;$

Wherein, the previous section that n is targeted sites; J is the next section that vehicle will arrive; for under situation k, vehicle is through running time between the required average station of section i, and l is the distance between vehicle and nearest website;

Described average velocity

calculate by following formula:

$\stackrel{\‾}{V}=\frac{1}{M}\underset{i=1}{\overset{M}{\mathrm{\Σ}}}{v}_i;$

Wherein, M is the GPS valid data number of this vehicle of collecting, v _ifor speed corresponding to each vehicle in gps data.

3. the bus arrival time real-time estimation method based on driving performance according to claim 1, is characterized in that: described vehicle x stops the delay time at stop calculate by following formula:

$\stackrel{\‾}{D_x}=D_b+\underset{i=j}{\overset{n}{\mathrm{\Σ}}}{t}_{(i,s)}^k;$

Wherein, the previous station that n is targeted sites; J is the downstream station nearest from vehicle current location;

for under situation k, vehicle is the mean residence time of i AT STATION;

Described each station is in the average dwell time of different periods

calculate by following formula:

D _b＝αT ₀+βT ₁+γT ₂；

Wherein: α, beta, gamma is time correction factor; T ₀pull in the time of sailing into for slowing down; T ₁for the boarding and alighting time that stops; T ₂for accelerating leaving from station rolling away from the time.

4. the bus arrival time real-time estimation method based on driving performance according to claim 1, is characterized in that: described vehicle is in signal lamp place stop delay time D _pcalculate by following steps:

S51: the public transit vehicle position location and the instantaneous velocity that utilize gps data Real-time Obtaining;

S52: determine that whether public transit vehicle is in signal lamp region;

S53: according to the transport condition of vehicle, calculate the residence time of vehicle in current traffic lights position by following formula:

S54: in the time that vehicle is met red light, from current location to the stop delay time sum of estimating all signal lamps station be

$D_p=\underset{p=1}{\overset{q}{\mathrm{\Σ}}}\underset{x=1}{\overset{M}{\mathrm{\Σ}}}P\left(x\right)T_{\mathrm{px}};$

S55: when vehicle is met when green light, start from the next signal lamp of current location to the stop delay time sum of estimating all signal lamps station be $D_p=\underset{p=2}{\overset{q}{\mathrm{\Σ}}}\underset{x=1}{\overset{M}{\mathrm{\Σ}}}P\left(x\right)T_{\mathrm{px}};$

S56: when vehicle is not during or not signal lamp region, from current location to the stop delay time sum of estimating all signal lamps station be $D_p=\underset{p=1}{\overset{q}{\mathrm{\Σ}}}\underset{x=1}{\overset{M}{\mathrm{\Σ}}}P\left(x\right)T_{\mathrm{px}};$

Wherein, in the time period that P (x) expression time interval length is t, there is the probability distribution in x car arriving signal lamp region, T _pxfor the random delay time at stop of vehicle parking at signal lamp p place, M is that the vehicle in signal lamp p place one-period arrives number.

Images