CN111491261B - Individual movement track extraction method based on intelligent card swiping data - Google Patents

Individual movement track extraction method based on intelligent card swiping data Download PDF

Info

Publication number
CN111491261B
CN111491261B CN202010333431.5A CN202010333431A CN111491261B CN 111491261 B CN111491261 B CN 111491261B CN 202010333431 A CN202010333431 A CN 202010333431A CN 111491261 B CN111491261 B CN 111491261B
Authority
CN
China
Prior art keywords
station
getting
probability
time
card swiping
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN202010333431.5A
Other languages
Chinese (zh)
Other versions
CN111491261A (en
Inventor
龚咏喜
王梦晗
金美含
王嘉琪
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Shenzhen Graduate School Harbin Institute of Technology
Original Assignee
Shenzhen Graduate School Harbin Institute of Technology
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Shenzhen Graduate School Harbin Institute of Technology filed Critical Shenzhen Graduate School Harbin Institute of Technology
Priority to CN202010333431.5A priority Critical patent/CN111491261B/en
Publication of CN111491261A publication Critical patent/CN111491261A/en
Application granted granted Critical
Publication of CN111491261B publication Critical patent/CN111491261B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
    • H04W4/02—Services making use of location information
    • H04W4/029—Location-based management or tracking services
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
    • H04W4/02—Services making use of location information
    • H04W4/024—Guidance services

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Traffic Control Systems (AREA)

Abstract

The invention discloses an individual movement track extraction method based on intelligent card swiping data, which comprises the following steps of: calculating a first probability of getting off from a first station to a grid and a second probability of getting on from the grid to a second station according to the intelligent card swiping data, the bus track data, the bus station and the line data; obtaining a joint probability according to the first probability and the second probability; calculating the number attraction of getting-off from the first station and the getting-off distance attenuation coefficient between the first station and the second station according to the intelligent card swiping data, the bus track data, the bus stations and the line data; obtaining a getting-off attraction according to the quantity attraction and the getting-off distance attenuation coefficient; and obtaining the comprehensive probability according to the joint probability and the getting-off attraction. Because the intelligent card swiping data and the related public transportation data are in the collection range for big and small cities in China, and the comprehensive probability is obtained by calculating the joint probability and the getting-off attraction, the individual moving track is extracted, and the method has high applicability and lower data processing difficulty.

Description

Individual movement track extraction method based on intelligent card swiping data
Technical Field
The invention relates to the technical field of traffic, in particular to an individual movement track extraction method based on intelligent card swiping data.
Background
In recent years, with the development of information technology, the data information amount is increased explosively, the data sources are more and more, and the data amount is also more and more huge. The activities of people in a city group are in a certain range, and the individual moving track obtained through the positioning equipment needs to be carried and opened at any time, so that the data acquisition and processing difficulty is high, and the privacy problem is easily caused when the data is obtained through the GPS positioning equipment in the mobile phone.
Accordingly, the prior art is yet to be improved and developed.
Disclosure of Invention
The invention provides an individual movement track extraction method based on intelligent card swiping data, aiming at solving the problem that the acquisition and processing difficulty of individual movement track data is high in the prior art.
The technical scheme adopted by the invention for solving the technical problem is as follows:
an individual movement track extraction method based on intelligent card swiping data comprises the following steps:
calculating a first probability of getting off from a first station to a grid and a second probability of getting on from the grid to a second station according to the intelligent card swiping data, the bus track data, the bus station and the line data; the first station is positioned in a circle which takes the second station as the center of the circle and takes the preset distance as the radius;
obtaining a joint probability according to the first probability and the second probability;
calculating the number attraction of getting-off from the first station and the attenuation coefficient of the getting-off distance between the first station and the second station according to the intelligent card swiping data, the bus track data, the bus stations and the line data;
obtaining a getting-off attraction according to the quantity attraction and the getting-off distance attenuation coefficient;
and obtaining a comprehensive probability according to the joint probability and the getting-off attraction.
The method for extracting the individual movement track based on the intelligent card swiping data is characterized in that the center of gravity of the grid is located in an elliptical area with the first station and the second station as focuses; the elliptical area is expressed by the following formula:
Figure BDA0002465768820000021
Figure BDA0002465768820000022
Figure BDA0002465768820000023
wherein the content of the first and second substances,
Figure BDA0002465768820000024
DM，Arespectively being a point M on the boundary of the elliptical area and an ith first site AiThe linear distance between the second stations A, DfFrom the ith first station A for passengersiThe distance walkable in the time between to the second station a,
Figure BDA0002465768820000025
is the ith first site AiDistance of road between second stops A, D0Is a predetermined distance.
The method for extracting the individual movement track based on the intelligent card swiping data is characterized in that the passenger moves from the ith first station AiDistance D walkable in time to second station AfComprises the following steps:
Df＝Tf×V；
Tf＝TA-TAi-Twait；
where V is the average walking speed of the passenger, TAArrival time, T, of the second station A vehicleAiIs the ith first site AiArrival time of the vehicle, TwaitIs the average waiting time.
The method for extracting the individual movement track based on the intelligent card swiping data comprises the following steps of:
Figure BDA0002465768820000026
the second probability is:
Figure BDA0002465768820000031
Figure BDA0002465768820000032
Figure BDA0002465768820000033
wherein the content of the first and second substances,
Figure BDA0002465768820000034
to arrive at the weight of the jth trellis,
Figure BDA0002465768820000035
is a weight starting from the jth grid, AreajIs the building area of the jth grid,
Figure BDA0002465768820000036
is the ith first site AiCenter of gravity H to jth meshjThe distance of the road (c) to the road,
Figure BDA0002465768820000037
as the center of gravity H of the second site A to the jth gridjN is the number of meshes in the elliptical area, and Σ is a summation symbol.
The individual movement track extraction method based on the intelligent card swiping data is characterized in that the joint probability is as follows:
Figure BDA0002465768820000038
the method for extracting the individual movement track based on the intelligent card swiping data comprises the following steps:
Figure BDA0002465768820000039
wherein the content of the first and second substances,
Figure BDA00024657688200000310
is the ith first station A in a preset time periodiThe right of attraction of the number of alighting,
Figure BDA00024657688200000311
is the ith first station A in a preset time periodiThe number of alighting persons, CsIs the ith first station A in a preset time periodiThe total number of people getting off the bus on the line; when the preset time interval is the early peak time interval, the later peak time interval
Figure BDA00024657688200000312
And CSCalculating early peak hours
Figure BDA00024657688200000313
When the preset time interval is late peak, the time interval is early peak
Figure BDA00024657688200000314
And CsCalculating late peak hours
Figure BDA00024657688200000315
When the preset time period is other than the early peak time period and the late peak time period, the preset time period
Figure BDA00024657688200000316
And CSCalculating said predetermined period
Figure BDA00024657688200000317
The individual movement track extraction method based on the intelligent card swiping data is characterized in that the descending distance attenuation coefficient is as follows:
Figure BDA0002465768820000041
wherein the content of the first and second substances,
Figure BDA0002465768820000042
is the ith first site AiThe road distance between the second stations a,
Figure BDA0002465768820000043
is the ith first site AiThe descending distance attenuation coefficient.
The intelligent card swiping data-based individual movement track extraction method comprises the following steps of:
Figure BDA0002465768820000044
the individual movement track extraction method based on the intelligent card swiping data is characterized in that the comprehensive probability is as follows:
Figure BDA0002465768820000045
wherein m is a first site AiThe number of the cells.
The method for extracting the individual movement track based on the intelligent card swiping data is characterized in that the preset distance is 2 kilometers.
Has the advantages that: because the intelligent card swiping data, the bus track data, the bus stop and the line data are in the collection range for big and small cities in China, the collection is easy; and the comprehensive probability is obtained by calculating the joint probability and the getting-off attraction, so that the individual movement track is extracted, and the data processing difficulty is reduced.
Drawings
Fig. 1 is a first flowchart of an individual movement trajectory extraction method based on smart card swiping data in the invention.
Fig. 2 is a second flowchart of the method for extracting the individual movement track based on the intelligent card swiping data in the invention.
Fig. 3 is a geographical schematic of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention clearer and clearer, the present invention is further described in detail below with reference to the accompanying drawings and examples. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
Referring to fig. 1-3, the present invention provides embodiments of an individual movement trajectory extraction method based on smart card swiping data.
The understanding of the travel state of people is of great significance to the understanding of urban activities, public transport trips are taken in a great proportion in daily travel of people, and the activities of groups in cities can be mostly mastered if the movement of people through a public transport system can be acquired. Therefore, the method for acquiring the individual moving tracks in the city by adopting the bus data (comprising the bus track data, the bus stops and the line data) is very important, the bus data is easy to acquire, and the processing difficulty of the individual moving tracks is reduced. The bus card swiping system widely used at present is mainly divided into an open system and a closed system. Wherein the closed system can provide the boarding station and the alighting station of individual passengers, and has more comprehensive information. The open system is used for the passengers to swipe cards only when getting on the bus, and the information reserved in the normal condition is only the information such as card swiping time, card swiping cost and the like. The invention relates to an individual movement track inference method developed by means of card swiping data, bus GPS track data and basic information such as stops and lines of an open system.
The invention mainly comprises two parts: an intelligent card swiping data acquisition part extracts the boarding stations of individuals; and a data processing section for estimating a departure place and a destination of the individual from the boarding data.
As shown in fig. 2, an individual movement trajectory extraction method based on smart card swiping data of the present invention includes the following steps:
step S100, calculating a first probability of getting off from a first station to a grid and a second probability of getting on from the grid to a second station according to intelligent card swiping data, bus track data, bus stations and line data; the first station is located in a circle with the second station as the center of circle and the preset distance as the radius.
Specifically, the bus data includes: bus (vehicle GPS) trajectory data; bus route data; bus stop data. The bus arrival schedule can be obtained through bus data, and the specific process is as follows:
1. map matching
Matching bus GPS track data with map of corresponding route
In the patent, all buses are assumed to be driven according to the set bus route. And matching the GPS track data of the bus with the corresponding line data of the bus through three factors of distance, direction and speed. And finding the position on the line corresponding to each GPS track point.
The line from the starting point to the end point is represented by a numerical value of 0 → 1, and the position value of each trace point on the line is obtained and is represented as the numerical value in 0 → 1.
② map matching of bus station data and corresponding line
And matching the bus stop data with the corresponding bus lines, and similarly obtaining the position value of each bus stop on the line, which is also expressed as the numerical value in 0 → 1.
2. Obtaining arrival time of vehicle
Writing code using python completes.
The algorithm logic is as follows: the following calculation is performed for each bus stop of each line:
firstly, finding out which two track points of a bus station are positioned in the same bus GPS track data.
Secondly, calculating the average speed by using the time difference and the distance difference between the GPS track points.
And thirdly, calculating the distance difference between the bus stop and the track point behind the bus stop, calculating the time difference according to the average speed, and determining the time of arriving at the bus stop based on the time of the track point.
Through the steps, a relatively accurate arrival schedule is obtained, and in general, the time for the passenger to get on the bus and swipe the card is after the arrival of the bus and before the bus arrives at the next station in the driving direction. Therefore, if the time of card swiping data is in the site A1And to site a2In the time interval, the boarding station of the passenger is considered as A1And (4) a station. The same is done with Python writing code.
As shown in fig. 1, the movement trajectory of the residents is estimated by using the public transportation data, that is, the movement trajectory is estimated by combining the riding records of the residents with the spatial position. The movement track of any individual is composed of three parts, namely a departure place of the first bus trip of the individual, an activity area of the middle arrival and an arrival place of the last trip. The principle that all individuals travel using only public transport was followed in the study.
The grid in the present application includes: a social management grid, a grid (shown in fig. 3, for simplicity, as a grid). We chose a social management grid as the basic unit of research when analyzing the passenger's activity area. Compared with the traditional grid, the social management grid has the advantages of clear boundary and undivided functions. But may be replaced by a grid division without a more detailed social management grid. The social management grids are uniformly numbered prior to analysis. The social management grid will be described below as an example.
The data required for analyzing different parts of the individual movement trajectory is different. The essential relationship of the data is the card swiping data of the individual for two consecutive boarding behaviors in time.
Determining possible activity area of individual by two times of card swiping data in succession
As shown in fig. 3, assume that the individual: station number of first-time ride line L1: s1，S2，S3，…，Sk，…，SlAnd the arrival time T of the vehicleS1，TS2，TS3，…，TSk，…，TSl(ii) a If S in L1 linekThe station is positioned in the range with the second station A as the center of circle and the radius of the second station A as the preset distanceIf so, the SkSelecting sites to form a site set, namely a first site AiAnd the number of the first stops is m, the second-time riding line L2 gets on the station A and the time T of the vehicle arriving at the station AA(ii) a A walking average speed V of the person; average waiting time Twait。
For example, the following steps are carried out: l1 has S in the line2，S3，S4The three stations are located within a circle having a radius of a predetermined distance from the center of the a station in the L2 line. That is, when a passenger takes the L1 line and transfers to the L2 line, the passenger may be at S2，S3，S4And (4) getting off at three stations. Then the arrival time of the vehicle according to the L1 line, S2，S3，S4The vehicle arrival times at the three stations are 8:30, 8:40, 8:50, and the vehicle arrival time at the L2 line a station is 9:15, respectively, then the passenger may be from S2，S3，S4One of the three stations gets off and transfers to station A, then for convenience of representation, S is used2，S3，S4Three sites configuration Ai-1，Ai，Ai+1。
A calculation step:
selecting all first sites A in a range with a radius of a preset distance and taking the second sites A as the circle centeriComputing site A and site AiFor road distance between stations
Figure BDA0002465768820000071
And (4) showing. Utilize A one by oneiStation and a station perform the following steps.
② determining that the passenger is in AiActive time T between station and A stationf：
Tf＝TA-TAi-Twait；
To distinguish from passengers at AiWhether the station transfers to A station or A stationiAnd (3) carrying out (life/work) activities after getting off the station, and then getting on and off the station A. It is easily understood that if it is a transfer, then TfSmaller, if it is a get-off activity, TfIs relatively large. Because the vehicle has a plurality of trips, TAThere are several, usually acquisition and TAiT at temporally adjacent time instantsA(of course, TA-TAi> 0), if Tf<0, that is, then AiThe space between the station car and the station A car is small, and the passenger is at the station AiAfter the station gets off, the driver cannot catch up with the vehicle of the station A and can only wait for the vehicle of the next shift of the station A. That is, T is ensuredf>0, of course TwaitCorresponding to the spacing between two adjacent cars of the A station (T can be adjusted)waitThe interval between two adjacent cars at the a station), that is, the passenger is at aiActive time T between station and A stationfTo judge that the passenger is at AiAnd after getting off the bus at the station, the bus is transferred or the bus is moved.
(iii) determining that the passenger is at AiDistance D capable of walking in activity time between station A and station Bf：
Df＝Tf×V；
First, compare
Figure BDA0002465768820000081
And DfJudging whether the reachable distance supports two riding behaviors or not, and updating the reachable distance:
Figure BDA0002465768820000082
if all A areiThe results obtained by site calculation are DfIf the card swiping action is 0, the transfer action is between the two card swiping actions. It is understood that when the above-mentioned material is used, the above-mentioned material can be
Figure BDA0002465768820000083
Due to TfIs subtracting TwaitThat is, TfCovering at least one pass with the passenger at TfAt least catch up with the vehicle at the A station, meaning that the passenger is transferring, DfThe configuration is 0. If it is not
Figure BDA0002465768820000084
Then it indicates that the passenger is at aiThe site is followed by an activity, DfIs not changed and is still Df。
Second, people generally choose to walk to reach their destination within a certain distance. We set this tolerance distance to 2km (i.e. the preset distance D)0) Thus the range of motion D for the passengerf’It is further concluded that:
Figure BDA0002465768820000091
that is, if the passenger is a transfer behavior, Df’0 if the passenger is AiThe passengers get on the station after getting off the station and the time for handling the station is shortened to the shortest, so that the passengers walk all the time and get on the station A after handling the station, and the region which can be reached by the passengers is AiStation and station A are in focus, DfAn ellipse with the major axis. If D isfLarger, over a predetermined distance D0That is, if the passenger stops between getting off and getting on, and does not walk all the time, D will bef’Arranged at a predetermined distance D0. For example, passengers are in the morning from AiAfter the station leaves the bus, the station works at night and gets on the bus to leave at the station A, and then the station D can be usedf’Arranged at a predetermined distance D0。
Fourthly, the passenger is at AiThe station gets off and then gets on the station A, and the maximum moving area between the stations is a moving distance Df’The value of (A) is the length of the major axisiStation and a-station are elliptical areas of focus.
The elliptical area expression is as follows:
Figure BDA0002465768820000092
in the formula, M: representing any point on the boundary of the elliptical region;
Figure BDA0002465768820000093
DM，A: respectively represent M and AiDistance between station and a station. By definition, the sum of the distance between any point on the ellipse and the two foci is a constant, and considering that the passenger may not walk all the time but may stop, any point in the elliptical area is likely to be where the passenger is to stay, and the major axis of the largest ellipse is a predetermined distance.
For example, as shown in FIG. 3, Twait＝10min，V＝1m/s。
For Ai-1In other words, the passenger is at Ai-1The active time between site and site a is: t isf(Ai-1-a) 35min, if ai-1The linear distance between the station and the station A is 1km, and passengers are driven from the station A by the linear distancei-1The time for the station to go to the a station is 1000 s-16.67 min, the time taken by the straight distance is removed, the passenger still has 35min-16.67 min-18.33 min, and the distance the passenger can travel in this 18.33min is 18.33min × 1 m/s-1.09 km, so the passenger is at ai-1The distance walkable in the activity time between the station and the station A is Df(Ai-1-a) ═ 1km +1.09km ═ 2.09 km. Due to the predetermined distance D0＝2km<2.09km, thus, for Ai-1For example, the elliptical area is:
Figure BDA0002465768820000094
Figure BDA0002465768820000101
for AiIn other words, the passenger is at AiThe active time between site and site a is: t isf(Ai-a) 25min, if aiThe linear distance between the station and the station A is 1.2km, and passengers are driven from the station A by the linear distanceiThe time for the station to go to the A station is 1200 s-20 min, except the time taken by the straight line distance, the passenger still has 25min-20 min-5 min, and the distance that the passenger can walk in the 5min is 5min x 1 m/s-0.3 km, so the passenger is in A stationiStation and station AThe distance that can be walked in the activity time between points is Df(Ai-1-a) ═ 1km +0.3km ═ 1.3 km. Due to the predetermined distance D0＝2km>1.3km, therefore, for AiFor example, the elliptical area is:
Figure BDA0002465768820000102
for Ai+1In other words, the passenger is at Ai+1The active time between site and site a is: t isf(Ai+1-a) 15min, if ai-1The linear distance between the station and the station A is 0.9km, and passengers are driven from the station A by the linear distancei+1The time taken for the station to travel to the a station is 900 s-15 min, except for the time taken by the straight distance, and the time taken for the passenger to travel by 15 min-0 min, the distance that the passenger can travel within this 0min is 0km, and thus, an elliptical region cannot be formed.
The social management grid with the gravity center covered by the activity range (oval area) is the possible activity area of the individual.
Of course, the elliptical area may cover multiple social management grids, requiring the computation of passenger slave AiThe first probability that the passenger gets off the station and arrives at any social management grid in the oval area is achieved, and the second probability that the passenger gets on the station A from any social management grid in the oval area is achieved.
Generally, the more people that can be accommodated within a grid, the greater the probability that an individual will depart from the grid, and conversely the more individuals will arrive at the grid. The volume may be expressed by the building area within the grid. The larger the building area, the more people can be accommodated. Meanwhile, the action of the person accords with the distance attenuation rule. Meanwhile, the essence of the distance decay law is that the interaction between geographical elements is related to the distance, and the interaction between geographical elements is inversely proportional to the square of the distance when other conditions are the same. The weight W for each grid can therefore be calculated using the following formulaj：
Figure BDA0002465768820000103
In the formula: wjA weight for departure/arrival for the jth social management grid; areajManaging the building area of the grid for the jth society;
Figure BDA0002465768820000111
is AjSite to jth social management grid center of gravity HjThe linear distance of (a).
And if the elliptical area is covered with n grids, the probability p of each grid is as follows:
Figure BDA0002465768820000112
specifically, the first probability is:
Figure BDA0002465768820000113
the second probability is:
Figure BDA0002465768820000114
Figure BDA0002465768820000115
Figure BDA0002465768820000116
wherein the content of the first and second substances,
Figure BDA0002465768820000117
to arrive at the weight of the jth trellis,
Figure BDA0002465768820000118
is a weight starting from the jth grid, AreajIs the building area of the jth grid,
Figure BDA0002465768820000119
is the ith first site AiCenter of gravity H to jth meshjThe linear distance of (a) is,
Figure BDA00024657688200001110
as the center of gravity H of the second site A to the jth gridjN is the number of meshes in the elliptical area, and Σ is the sum sign.
And S200, obtaining a joint probability according to the first probability and the second probability.
Specifically, for the same social management grid, the two probabilities are multiplied to obtain a joint probability, namely, the joint probability represents that the individual is from AiThe probability that the station gets off to a social management grid and gets on from the grid to the A station is used as the joint probability pjAnd (4) showing. The formula is as follows:
Figure BDA00024657688200001111
step S300, calculating the number attraction of getting off from the first station and the getting-off distance attenuation coefficient between the first station and the second station according to the intelligent card swiping data, the bus track data, the bus stations and the line data.
Specifically, we select all A's within a predetermined distance from the A site radiusiStations, but passengers from these AiThe probability of getting off the station is not the same. Therefore, the possibility of getting off an individual from each bus stop needs to be measured through getting off attraction of the bus stop.
When considering the getting-off attraction right of the bus stop, the bus stop consists of two parts. The number attraction represented by the number of the passengers getting off at each bus stop on the same line is partially considered due to the limitation of a bus card swiping system, and the passengers getting on the bus are represented by the data. The other part is that the distance between the bus station and the passenger getting-on station for the second time is related and is characterized by a falling distance attenuation coefficient.
1. Number attraction
Generation amount of bus stops: the number of passengers getting on from the bus stop; attraction of bus stops: and calculating the getting-off attraction of the bus stop based on the attraction amount by the number of the passengers getting off at the bus stop. Based on the rule that urban residents use public transport means, the occurrence amount and the attraction amount of each bus stop can be found to have correlation, and therefore the getting-off attraction of the bus stops can be calculated by using the getting-on data.
Secondly, in the non-morning and evening peak time periods, the occurrence amount (the number of people getting on the bus from the place) and the attraction amount (the number of people getting off the bus from the place) of a certain bus stop are roughly balanced. In the morning and evening peak hours, the generation amount and the suction amount of a part of nearby bus stops with single functions are unbalanced, but the generation amount in the morning peak hours, the suction amount in the evening peak hours, the generation amount in the evening peak hours and the suction amount in the morning peak hours are closer. Therefore, the morning and evening peak boarding data can be exchanged to calculate the getting-off attraction of the bus stop in another time period.
And thirdly, calculating a getting-off attraction for each bus stop of the upward stop and the downward stop of each bus line respectively. And three different values of the early peak time (6:00-9:30), the late peak time (16:30-20:00) and other time (9:30-16:30, 20:00-23:00) are respectively calculated.
Therefore, taking a certain line as an example, calculate AiThe number of alights of the station:
Figure BDA0002465768820000131
in the formula:
Figure BDA0002465768820000132
is the ith first site AiThe number of alights of getting-off;
Figure BDA0002465768820000133
indicating the ith first site a in rush houriNumber of alighting persons, e.g. when departure from the head stationDuring the early peak period, the line goes up (down) at AiThe number of people getting off the bus at the station can be within the late peak time at the departure time of the first station, and the line goes up (down) at AiThe number of people getting on the bus at the station; cSIndicating the ith first site a in rush houriCounting the total number of people getting off in the line where the station is located and counting the number of the first station AiAll stations in the route (e.g., all stations of the route are S)1，S2，S3，…，Sk，…，SlFor example, S3Station is actually AiStop) gets the total number, and the departure time of the first stop is in the early peak time, the total number of people who get off the line in an ascending (descending) way can be replaced by the total number of people who get on the line in an ascending (descending) way when the departure time of the first stop is in the late peak time.
Similarly, take a line as an example, late peak hours, AiThe getting-off attraction of the station is calculated by using the data of the departure time of the first station in the early peak time; other periods of time, AiAnd the getting-off attraction of the station is calculated by using the data of the departure time of the first station in other time periods.
2. Damping coefficient of departure distance
Is at AiUnder the influence of the distance between the station and the station A, the method also has certain influence on the selection of the get-off station by individuals. Thus making use of AiAnd calculating a getting-off distance attenuation coefficient according to the distance between the station and the station A.
The formula is as follows:
Figure BDA0002465768820000134
in the formula:
Figure BDA0002465768820000135
is represented by AiThe getting-off distance attenuation coefficient of the station;
Figure BDA0002465768820000136
is represented by AiRoad distance of station and a station.
And S400, obtaining the getting-off attraction according to the number attraction and the getting-off distance attenuation coefficient.
Specifically, the getting-off attraction of the station is calculated in consideration of the number attraction and the getting-off distance attenuation coefficient. The getting-off attraction right is as follows:
Figure BDA0002465768820000141
and S500, obtaining a comprehensive probability according to the joint probability and the getting-off attraction.
Calculating the sum of the joint probabilities by combining the lower vehicle attraction, normalizing, and determining the comprehensive probability P of each gridj：
Figure BDA0002465768820000142
Normalizing the product of the joint probability and the alighting attraction, i.e. meaning
Figure BDA0002465768820000143
It is to be noted that the present invention has the following effects.
1. The invention has lower requirement on the data, and is carried out on the basic bus card swiping data, the bus GPS track and the bus station line data, and the data are in the acquisition range for big and small cities in China. Therefore, the applicability is wide.
2. The invention infers from the site to the region from the space-time perspective, thereby analyzing the individual movement track. At present, most of public transportation data are utilized to analyze individual movement, and the individual movement focuses on the inference of an getting-on station and a getting-off station of an individual, but the part from the station to an area is also relatively concerned. Meanwhile, the social management grid can be used as a basic unit for reasonably researching the movement of people.
3. The method can be used as a basic content, and can also infer the problems of the probability of the getting-off station of the individual and the like on the basis of the individual track inference.
In summary, the method for extracting the individual movement track based on the intelligent card swiping data provided by the invention comprises the following steps: calculating a first probability of getting off from a first station to a grid and a second probability of getting on from the grid to a second station according to the intelligent card swiping data, the bus track data, the bus station and the line data; the first station is positioned in a circle which takes the second station as the center of the circle and takes the preset distance as the radius; obtaining a joint probability according to the first probability and the second probability; calculating the number attraction of getting-off from the first station and the attenuation coefficient of the getting-off distance between the first station and the second station according to the intelligent card swiping data and the bus data; obtaining a getting-off attraction according to the quantity attraction and the getting-off distance attenuation coefficient; and obtaining the comprehensive probability according to the joint probability and the getting-off attraction. Because the intelligent card swiping data and the related public transportation data are in the collection range for big and small cities in China, the collection is easy; and the comprehensive probability is obtained by calculating the joint probability and the getting-off attraction, so that the method has high applicability and low data processing difficulty by extracting the individual movement track.
It is to be understood that the invention is not limited to the examples described above, but that modifications and variations may be effected thereto by those of ordinary skill in the art in light of the foregoing description, and that all such modifications and variations are intended to be within the scope of the invention as defined by the appended claims.

Claims (10)

1. An individual movement track extraction method based on intelligent card swiping data is characterized by comprising the following steps:
calculating a first probability of getting off from a first station to a grid and a second probability of getting on from the grid to a second station according to the intelligent card swiping data, the bus track data, the bus station and the line data; the first station is positioned in a circle which takes the second station as the center of the circle and takes the preset distance as the radius;
obtaining a joint probability according to the first probability and the second probability;
calculating the number attraction of getting-off from the first station and the attenuation coefficient of the getting-off distance between the first station and the second station according to the intelligent card swiping data, the bus track data, the bus stations and the line data;
obtaining a getting-off attraction according to the quantity attraction and the getting-off distance attenuation coefficient;
obtaining a comprehensive probability according to the joint probability and the getting-off attraction; the comprehensive probability refers to the probability of the passenger reaching the grid when the station and the grid are comprehensively considered based on the space-time angle.
2. The method for extracting individual movement trajectory based on smart card swiping data according to claim 1, wherein the center of gravity of the grid is located in an elliptical area with the first site and the second site as the focus; the elliptical area is expressed by the following formula:
Figure FDA0003426023350000011
Figure FDA0003426023350000012
Figure FDA0003426023350000013
wherein the content of the first and second substances,
Figure FDA0003426023350000014
respectively being a point M on the boundary of the elliptical area and an ith first site AiThe linear distance between the second stations A, DfFrom the ith first station A for passengersiThe time formed between the time of getting off to the time of getting on at the second station AThe interval is divided into a plurality of sections, the distance which can be walked in the activity time of the average waiting time is removed,
Figure FDA0003426023350000015
is the ith first site AiDistance of road between second stops A, D0Is a predetermined distance.
3. The method for extracting individual movement track based on intelligent card swiping data according to claim 2, wherein the passenger is from the ith first site AiThe time period formed from the time of getting off to the time of getting on at the second station A and the distance D which can be walked in the activity time except the average waiting timefComprises the following steps:
Df＝Tf×V；
Tf＝TA-TAi-Twait；
where V is the average walking speed of the passenger, TAArrival time, T, of the second station A vehicleAiIs the ith first site AiArrival time of the vehicle, TwaitIs the average waiting time, TfFor passengers at first station AiThe time period formed between the time of getting off the train and the time of getting on the train at the second station A is divided by the activity time of the average waiting time.
4. The method for extracting the individual movement track based on the intelligent card swiping data according to claim 2, wherein the first probability is as follows:
Figure FDA0003426023350000021
the second probability is:
Figure FDA0003426023350000022
Figure FDA0003426023350000023
Figure FDA0003426023350000024
wherein the content of the first and second substances,
Figure FDA0003426023350000025
to arrive at the weight of the jth trellis,
Figure FDA0003426023350000026
is a weight starting from the jth grid, AreajIs the building area of the jth grid,
Figure FDA0003426023350000027
is the ith first site AiCenter of gravity H to jth meshjThe distance of the road (c) to the road,
Figure FDA0003426023350000028
as the center of gravity H of the second site A to the jth gridjN is the number of meshes in the elliptical area, and Σ is a summation symbol.
5. The method for extracting individual movement track based on intelligent card swiping data according to claim 4, wherein the joint probability is as follows:
Figure FDA0003426023350000031
6. the method for extracting the individual movement track based on the intelligent card swiping data according to claim 5, wherein the number attraction is as follows:
Figure FDA0003426023350000032
wherein the content of the first and second substances,
Figure FDA0003426023350000033
is the ith first station A in a preset time periodiThe right of attraction of the number of alighting,
Figure FDA0003426023350000034
is the ith first station A in a preset time periodiThe number of alighting persons, CsIs the ith first station A in a preset time periodiThe total number of people getting off the bus on the line; when the preset time interval is the early peak time interval, the later peak time interval
Figure FDA0003426023350000035
And CsCalculating early peak hours
Figure FDA0003426023350000036
When the preset time interval is late peak, the time interval is early peak
Figure FDA0003426023350000037
And CsCalculating late peak hours
Figure FDA0003426023350000038
When the preset time period is other than the early peak time period and the late peak time period, the preset time period
Figure FDA0003426023350000039
And CsCalculating said predetermined period
Figure FDA00034260233500000310
7. The method for extracting an individual movement track based on intelligent card swiping data according to claim 6, wherein the getting-off distance attenuation coefficient is as follows:
Figure FDA00034260233500000311
wherein the content of the first and second substances,
Figure FDA00034260233500000312
is the ith first site AiThe road distance between the second stations a,
Figure FDA00034260233500000313
is the ith first site AiThe descending distance attenuation coefficient.
8. The method for extracting the individual movement track based on the intelligent card swiping data as claimed in claim 7, wherein the getting-off attraction is as follows:
Figure FDA00034260233500000314
9. the method for extracting the individual movement track based on the intelligent card swiping data according to claim 8, wherein the comprehensive probability is as follows:
Figure FDA00034260233500000315
wherein m is a first site AiThe number of the cells.
10. The method for extracting an individual moving track based on intelligent card swiping data according to any one of claims 1 to 9, wherein the predetermined distance is 2 kilometers.
CN202010333431.5A 2020-04-24 2020-04-24 Individual movement track extraction method based on intelligent card swiping data Active CN111491261B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202010333431.5A CN111491261B (en) 2020-04-24 2020-04-24 Individual movement track extraction method based on intelligent card swiping data

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202010333431.5A CN111491261B (en) 2020-04-24 2020-04-24 Individual movement track extraction method based on intelligent card swiping data

Publications (2)

Publication Number Publication Date
CN111491261A CN111491261A (en) 2020-08-04
CN111491261B true CN111491261B (en) 2022-03-01

Family

ID=71800257

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202010333431.5A Active CN111491261B (en) 2020-04-24 2020-04-24 Individual movement track extraction method based on intelligent card swiping data

Country Status (1)

Country Link
CN (1) CN111491261B (en)

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101615340A (en) * 2009-07-24 2009-12-30 北京工业大学 Real-time information processing method in the bus dynamic dispatching
CN107545730A (en) * 2017-09-08 2018-01-05 哈尔滨工业大学 A kind of website based on Based on Bus IC Card Data is got on or off the bus passenger's number estimation method
CN109035770A (en) * 2018-07-31 2018-12-18 上海世脉信息科技有限公司 The real-time analyzing and predicting method of public transport passenger capacity under a kind of big data environment
CN110084442A (en) * 2019-05-16 2019-08-02 重庆大学 A kind of method of joint public transport and the progress passenger flow OD calculating of rail traffic brushing card data

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US9887566B2 (en) * 2015-12-28 2018-02-06 International Business Machines Corporation System for charging mobile device using an ad-hoc infrastructure with energy harvesting capabilities
CN105809292B (en) * 2016-03-21 2019-11-26 广州地理研究所 Bus IC card passenger getting off car website projectional technique

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101615340A (en) * 2009-07-24 2009-12-30 北京工业大学 Real-time information processing method in the bus dynamic dispatching
CN107545730A (en) * 2017-09-08 2018-01-05 哈尔滨工业大学 A kind of website based on Based on Bus IC Card Data is got on or off the bus passenger's number estimation method
CN109035770A (en) * 2018-07-31 2018-12-18 上海世脉信息科技有限公司 The real-time analyzing and predicting method of public transport passenger capacity under a kind of big data environment
CN110084442A (en) * 2019-05-16 2019-08-02 重庆大学 A kind of method of joint public transport and the progress passenger flow OD calculating of rail traffic brushing card data

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
智能公交系统背景下的公交调度优化研究;穆礼彬;《中国优秀硕士学位论文全文数据库》;20140131;全文 *

Also Published As

Publication number Publication date
CN111491261A (en) 2020-08-04

Similar Documents

Publication Publication Date Title
CN109308546B (en) Method and system for predicting bus trip get-off station of passenger
CN109035770B (en) Real-time analysis and prediction method for bus passenger capacity in big data environment
CN109034566B (en) A kind of intelligent dispatching method and device based on passenger flow above and below bus station
CN106652434B (en) A kind of bus dispatching method coordinated based on rail traffic
CN109102114B (en) Bus trip getting-off station estimation method based on data fusion
CN105702035B (en) A kind of method for complexity of being ridden using history public transport data assessment
CN109903553B (en) Multi-source data mining bus station identification and inspection method
CN112036757B (en) Mobile phone signaling and floating car data-based parking transfer parking lot site selection method
CN112784000B (en) Passenger searching method based on taxi track data
CN108364464A (en) A kind of public transit vehicle hourage modeling method based on probabilistic model
CN111161531A (en) Method for judging forward route, forward route and station passing based on bus-mounted terminal
CN112085249A (en) Customized bus route planning method based on reinforcement learning
CN115527369A (en) Large passenger flow early warning and evacuation method under large-area delay condition of airport hub
CN113079463A (en) Tourist attraction tourist travel activity identification method based on mobile phone signaling data
CN112329989A (en) Bus route planning method and device based on cloud computing and storage medium
CN108197078B (en) Method for calculating bus section passenger flow based on bus IC card data
CN111491261B (en) Individual movement track extraction method based on intelligent card swiping data
Lam et al. Prediction of bus arrival time using real-time on-line bus locations
Kostakos Using Bluetooth to capture passenger trips on public transport buses
CN116542404A (en) Prediction method for continuous transfer time of station-supporting passengers of passenger transportation hub station
Sun et al. Characterizing multimodal transfer time using smart card data: The effect of time, passenger age, crowdedness and collective pressure
CN113361885B (en) Dual-target urban public transportation benefit evaluation method based on multi-source data
CN111339159B (en) Analysis mining method for one-ticket public transport data
JP7425680B2 (en) Navigation device and navigation method
CN111931968B (en) Method for optimizing public bicycle station layout by using mobile phone GPS data

Legal Events

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