CN114550459A - Method for accurately predicting bus arrival based on big data and dynamic multidimensional - Google Patents

Method for accurately predicting bus arrival based on big data and dynamic multidimensional Download PDF

Info

Publication number
CN114550459A
CN114550459A CN202210436807.4A CN202210436807A CN114550459A CN 114550459 A CN114550459 A CN 114550459A CN 202210436807 A CN202210436807 A CN 202210436807A CN 114550459 A CN114550459 A CN 114550459A
Authority
CN
China
Prior art keywords
time
station
vehicle
arrival
data
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202210436807.4A
Other languages
Chinese (zh)
Inventor
刘晟
范汇涛
李刚
刘跃
周慧博
孙龙
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.)
Nanjing Intelligent Transportation Information Co ltd
Original Assignee
Nanjing Intelligent Transportation Information Co ltd
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 Nanjing Intelligent Transportation Information Co ltd filed Critical Nanjing Intelligent Transportation Information Co ltd
Priority to CN202210436807.4A priority Critical patent/CN114550459A/en
Publication of CN114550459A publication Critical patent/CN114550459A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G08—SIGNALLING
    • G08G—TRAFFIC CONTROL SYSTEMS
    • G08G1/00—Traffic control systems for road vehicles
    • G08G1/01—Detecting movement of traffic to be counted or controlled
    • G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
    • G08G1/0125—Traffic data processing
    • G—PHYSICS
    • G08—SIGNALLING
    • G08G—TRAFFIC CONTROL SYSTEMS
    • G08G1/00—Traffic control systems for road vehicles
    • G08G1/01—Detecting movement of traffic to be counted or controlled
    • G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
    • G08G1/0125—Traffic data processing
    • G08G1/0129—Traffic data processing for creating historical data or processing based on historical data
    • G—PHYSICS
    • G08—SIGNALLING
    • G08G—TRAFFIC CONTROL SYSTEMS
    • G08G1/00—Traffic control systems for road vehicles
    • G08G1/01—Detecting movement of traffic to be counted or controlled
    • G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
    • G08G1/0137—Measuring and analyzing of parameters relative to traffic conditions for specific applications
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
    • Y02T10/00—Road transport of goods or passengers
    • Y02T10/10—Internal combustion engine [ICE] based vehicles
    • Y02T10/40—Engine management systems

Landscapes

  • Chemical & Material Sciences (AREA)
  • Analytical Chemistry (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Traffic Control Systems (AREA)

Abstract

The invention discloses a method for accurately predicting bus arrival based on big data and dynamic multidimensional, belonging to the technical field of intelligent bus application .

Description

Method for accurately predicting bus arrival based on big data and dynamic multidimensional
Technical Field
The invention belongs to the technical field of intelligent public transport application, and particularly relates to a method for accurately predicting arrival of a public transport vehicle based on big data and dynamic multi-dimension.
Background
In the current bus arrival prediction method, the method of calculating the arrival time based on the real-time Gps of the vehicle and the vehicle speed provided by the vehicle-mounted terminal has the problems of low calculation precision, large cross-station error, no response mechanism to sudden situations and the like;
the existing electronic stop board is not popularized to each city, even if some cities run, the trial result is not ideal, the service of the electronic stop board does not fully consider various requirements of citizens for riding, the problems of unreasonable design, abnormal display content, wrong display or no display are existed, and the electronic stop board can not be correctly updated due to the fact that data can not be obtained due to communication interruption, data loss or power supply cut-off and the like;
the passengers cannot obtain reliable results to reasonably plan a trip scheme and reduce waiting time, so that a method for accurately predicting the arrival of the bus based on big data and dynamic multi-dimension is needed to solve the existing problems.
Disclosure of Invention
The invention aims to provide a method for accurately predicting the arrival time of a bus based on big data and dynamic multi-dimension, so as to solve the problem that the arrival time of the bus is not accurate due to prediction.
In order to achieve the purpose, the invention provides the following technical scheme: a method for accurately predicting bus arrival based on big data and dynamic multi-dimension comprises the following steps:
step 1, collecting historical data of buses, and establishing a bus arrival data warehouse of a bus route at different peak sections;
step 2, adopting a device for acquiring vehicle arrival information and vehicle departure information to acquire the vehicle arrival information and the vehicle departure information, screening unqualified passenger flow data, and inputting the qualified passenger flow data into a data warehouse;
step 3, identifying the use weight, analyzing and calculating the speed weight of the passenger flow to the vehicle on each line, analyzing and calculating the weight of the passenger flow to the station residence time, and taking the weight as an input factor of a measuring and calculating model;
step 4, establishing a speed curve chart of the bus arrival warehouse of each peak section, setting a forward movement fluctuation value and a backward movement fluctuation value, and comparing the forward movement fluctuation value and the backward movement fluctuation value with the real-time speed;
step 5, cleaning and sorting bus gps deviation correction and vehicle track matching data, associating the bus gps deviation correction and vehicle track matching data with passenger flow weight, vehicle alarm weight and speed curve graph interval conditions, using the bus gps deviation correction and vehicle track matching data as input factors of a bus route track vehicle arrival measuring and calculating model, and calculating the time of a vehicle from a first station;
and 6, establishing a training set, and predicting the arrival time of the vehicle to the subsequent station by using the training optimal model.
Preferably, the historical data of the public transport vehicles comprises: the bus route departure time, the interval time between adjacent stations, the arrival time and the traffic light waiting time;
the bus departure data acquisition data source is connected with the dispatching system;
the arrival time and the interval time between adjacent stations are obtained by analyzing station entrance and exit data uploaded by a bus-mounted terminal;
the traffic light data acquisition interface is connected with the third-party platform interface.
Preferably, the model of the bus arrival data warehouse comprises:
a next station time prediction model for predicting a time when the vehicle is away from the next station;
the inter-station time prediction model is used for predicting the time between the stations in the subsequent interval of the vehicle;
and the N-station time prediction model is used for predicting N-station time after the distance, wherein the N-station time after the distance is the sum of the time prediction of the next station and the time of the subsequent station at intervals.
Preferably, the device for acquiring the arrival information and the departure information of the vehicle comprises a vehicle-mounted terminal installed on the bus, and the vehicle-mounted terminal is used for acquiring the arrival passenger flow, the departure passenger flow, the first door opening time, the last door closing time and the vehicle stop time; the vehicle-mounted terminal is connected with the camera and the sensor, and the sensor identifies human face behaviors to judge the time of getting on or off the bus.
Preferably, the screening step of the unqualified passenger flow data comprises the following steps:
deleting data of real-time passenger flow + upper passenger flow-lower passenger flow < 0;
deleting data with the last passenger flow-the next passenger flow exceeding 100;
and deleting data with the last passenger flow or the next passenger flow exceeding 100.
Preferably, in step 5, the calculated time of the first station is matched with the arrival time of the similar scene of the historical data warehouse, the time of the second station is calculated within a controllable range, if the input result is not controllable, the time is recorded and matched after waiting for arrival in real time, and if the time difference with the arrival in real time is greater than a set value, the abnormal record is notified to the administrator to change the line or time period influence factor.
Preferably, the step of analyzing the influence of the passenger flow volume on the speed of the vehicle on different lines comprises:
calculating the sectional running speeds of different vehicles on different lines according to the historical gps data of the vehicles;
calculating the passenger flow of different sections of different lines through historical station entrance and exit collection;
and obtaining a linear relation through the speeds of different time segments and the passenger flows of different time segments.
Preferably, the first and second liquid crystal materials are,
predicted time to next station T1:
Figure 443692DEST_PATH_IMAGE001
subsequent interval station time prediction T2:
Figure 624137DEST_PATH_IMAGE002
arrival time T = T1+ T2;
Figure 692587DEST_PATH_IMAGE003
wherein the content of the first and second substances,
Figure 283975DEST_PATH_IMAGE004
the mileage between the vehicle and the nearest platform is taken as,
Figure 805086DEST_PATH_IMAGE005
the next station mileage of the vehicle from the nearest station,
Figure 241752DEST_PATH_IMAGE006
the real-time speed of the vehicle is obtained,
Figure 695867DEST_PATH_IMAGE007
in order to influence the factors for the arrival dwell time,
Figure 208888DEST_PATH_IMAGE008
waiting for an impact factor for a traffic light;
Figure 732142DEST_PATH_IMAGE009
the mileage between the nearest platform of the vehicle and the next platform of the vehicle; m is the serial number of the station of the calculation station, m +1 is the serial number of the next station of the calculation station,
Figure 988811DEST_PATH_IMAGE010
the distance between the station and the next station.
Preferably, the message format of the bus arrival speed at the peak section is as follows:
00 VVVV _ WW _ X _ Y _ Z, wherein V represents line coding; w represents a service; x represents a date; y represents a peak section; z represents the station code.
Preferably, in step 5, the method for cleaning and arranging the deviation rectification of the public transport gps comprises the following steps: calculating the preset time T by the GPS coordinate valueXDynamic mileage C of inner vehicle from starting stationsAnd dynamic mileage C of vehicle from end stationmIf the time T is presetX Increase in the value of (C) and dynamic range CS By decreasing the value of, or by a predetermined time TX Numerical value of (1) increase dynamic mileage Cm Numerical value ofAnd if the distance is increased progressively, eliminating the data exceeding the preset value of the distance road network offset.
The invention has the technical effects and advantages that: the method for accurately predicting the arrival of the bus based on the big data and the dynamic multi-dimension comprises the steps of establishing a model which influences the real-time running speed of the bus based on multiple factors such as real-time Gps, real-time vehicle-mounted passenger flow, traffic light waiting and vehicle alarming, carrying out data training based on different periods of different lines to obtain the accurate prediction of the arrival time of the bus at any point from a subsequent station, correcting the arrival predicted time of the bus and the actual arrival time by a mechanism when the bus enters or exits, and realizing the mechanism of automatically changing the weight of the influence after the correction is triggered for multiple times The method comprises the steps of accurately calculating the running time of a vehicle and reducing predicted errors by establishing bus arrival warehouse speed curve graphs of different peak sections, setting forward movement fluctuation values and backward movement fluctuation values and comparing the forward movement fluctuation values and the backward movement fluctuation values with real-time vehicle speed, correcting the predicted time of the bus arrival and the actual arrival time by a mechanism when the bus arrival and the actual arrival time go in and go out, and realizing a mechanism for automatically changing the weight of influence after the correction is triggered for multiple times, thereby improving the accuracy of prediction; the normal work of the electronic stop board is guaranteed by using a low-cost and effective means, and the speed prediction of different road sections is increased; the stop reporting algorithm of the electronic stop board is optimized; vehicle loss handling in special cases: analyzing the influence of traffic lights and passenger flow on arrival of vehicles; establishing a model for influencing the real-time running speed of the vehicle based on multiple factors such as real-time Gps, real-time vehicle-mounted passenger flow, traffic light waiting, vehicle alarming and the like, and performing data training based on different lines and different time periods to obtain accurate prediction of the arrival time of the vehicle at any point from a subsequent station; through the multi-dimensional bus route departure time, the adjacent station interval time, the arrival time and the traffic light waiting time, the connection between the station reporting data and the vehicles is tighter, and the station reporting minutes are more accurate.
Drawings
FIG. 1 is a flow chart of the present invention;
FIG. 2 is a graph of a data factor distribution according to the present invention;
FIG. 3 is a schematic diagram of model transformation according to the present invention;
FIG. 4 is a schematic illustration of the arrival prediction of the present invention;
fig. 5 is a bus arrival speed curve diagram of the invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
The invention provides a method for accurately predicting bus arrival based on big data and dynamic multi-dimension as shown in figures 1 and 4, which comprises the following steps:
step 1, collecting historical data of buses, and establishing a bus arrival data warehouse of a bus route at different peak sections; as shown in fig. 2, the history data of the public transportation vehicles includes: the bus route departure time, the interval time between adjacent stations, the arrival time and the traffic light waiting time;
in this embodiment, the model of the bus arrival data warehouse includes:
a next station time prediction model for predicting a time when the vehicle is away from the next station;
the inter-station time prediction model is used for predicting the time between the stations in the subsequent interval of the vehicle;
and the N-station time prediction model is used for predicting N-station time after the distance, wherein the N-station time after the distance is the sum of the time prediction of the next station and the time of the subsequent station at intervals.
The bus departure data acquisition data source is connected with the dispatching system;
the arrival time and the interval time between adjacent stations are obtained by analyzing station entrance and exit data uploaded by a bus-mounted terminal;
the traffic light data acquisition interface is connected with the third-party platform interface.
Step 2, adopting a device for acquiring vehicle arrival information and vehicle departure information to acquire the vehicle arrival information and the vehicle departure information, screening unqualified passenger flow data, and inputting the qualified passenger flow data into a data warehouse; the obtaining of the arrival information and the departure information of the bus comprises the steps that a vehicle-mounted terminal installed on the bus acquires data such as arrival and arrival passenger flow, departure passenger flow, first door opening, last door closing time, vehicle stop time and the like; the vehicle-mounted terminal is connected with the camera and the sensor, and the sensor identifies the human face behavior to judge the getting-on/off time.
Step 3, identifying the use weight, using the time period and the line weight, analyzing and calculating the influence weight of the passenger flow on the speed of the vehicle on different lines and analyzing and calculating the influence weight of the passenger flow on the station residence time, and taking the influence weights as the input factors of the measuring and calculating model; the step of analyzing the influence of the passenger flow on the speed of the vehicle on different lines comprises the following steps:
calculating the sectional running speeds of different vehicles on different lines according to the historical gps data of the vehicles;
calculating the passenger flow of different sections of different lines through historical station entrance and exit collection;
obtaining a linear relation through the speeds of different time segments and the passenger flows of different time segments;
predicted time to next station T1:
Figure 284051DEST_PATH_IMAGE001
subsequent interval station time prediction T2:
Figure 233552DEST_PATH_IMAGE002
arrival time T = T1+ T2;
Figure 729256DEST_PATH_IMAGE003
wherein the content of the first and second substances,
Figure 38883DEST_PATH_IMAGE004
the mileage between the vehicle and the nearest platform is taken as,
Figure 936432DEST_PATH_IMAGE005
the next station mileage of the vehicle from the nearest station,
Figure 791256DEST_PATH_IMAGE006
the real-time speed of the vehicle is obtained,
Figure 554681DEST_PATH_IMAGE007
in order to influence the factors for the arrival dwell time,
Figure 153153DEST_PATH_IMAGE008
waiting for an impact factor for a traffic light;
Figure 905208DEST_PATH_IMAGE009
the mileage between the nearest platform of the vehicle and the next platform of the vehicle; m is the serial number of the station of the calculation station, m +1 is the serial number of the next station of the calculation station,
Figure 711359DEST_PATH_IMAGE010
the distance between the station and the next station.
Step 4, establishing a speed curve chart of the bus arrival warehouse of each different peak section, setting a forward movement fluctuation value and a backward movement fluctuation value, and comparing the forward movement fluctuation value and the backward movement fluctuation value with the real-time speed; in the specific road section, when the vehicle speed is too fast or too slow, a forward offset value or a backward offset value can be manually configured;
step 5, as shown in fig. 3, cleaning and sorting bus gps deviation correction and vehicle track matching data, associating the bus gps deviation correction and vehicle track matching data with passenger flow weight, vehicle alarm weight and speed curve graph interval conditions, using the bus gps deviation correction and vehicle track matching data as input factors of a bus arrival measuring model of a bus route track, and calculating the time of a vehicle from a first station; matching the calculated time of the first station with the arrival time of the similar scene of the historical data warehouse, calculating the time of the second station within a controllable range, if the input result is uncontrollable, recording and waiting for the arrival in real time and then matching, and if the time difference with the arrival in real time is greater than a set value, recording an abnormal condition and informing an administrator to change the line or time period influence factor; the screening step of the unqualified passenger flow data comprises the following steps:
deleting data of real-time passenger flow + upper passenger flow-lower passenger flow < 0;
deleting data with the last passenger flow-the next passenger flow exceeding 100;
and deleting data with the last passenger flow or the next passenger flow exceeding 100.
And 6, establishing a training set, and predicting the arrival time of the vehicle to the subsequent station by using the training optimal model.
The message format of the bus arrival speed of the peak section is as follows:
00 VVVV _ WW _ X _ Y _ Z, wherein V represents line coding; w represents a service; x represents a date; y represents a peak section; z represents a station code;
in step 5, the method for cleaning and arranging the public transport gps deviation correction comprises the following steps: calculating the preset time T by the GPS coordinate valueXDynamic mileage C of inner vehicle from starting stationsAnd dynamic mileage C of vehicle from end stationmIf the time T is presetX Increase in the value of (C) and dynamic range CS By decreasing the value of, or by a predetermined time TX Numerical value of (1) increase dynamic mileage Cm In the embodiment, the existing bus positioning mostly adopts Gps or beidou positioning at present, and the signal blockage can cause the positioning offset phenomenon, so that the offset needs to be processed; the vehicle generally starts from a starting station to an end station, the calculated distance from the starting station to the vehicle is larger and smaller, and the distance from the end station to the vehicle is smaller and smaller. Therefore, if the distance calculated by the new gps coordinate is larger than the last time, the offset is generated, and the offset needs to be discarded;
the bus arrival time prediction method based on big data and dynamic multidimensional precision comprises the steps of establishing a model based on real-time Gps, real-time vehicle-mounted passenger flow, traffic light waiting, vehicle alarming and other multiple factors to influence the real-time running speed of a vehicle, carrying out data training based on different periods of different lines to obtain the precision prediction of the arrival time of the vehicle at any point from a subsequent station, correcting the arrival prediction time of the vehicle and the actual arrival time of the vehicle when the vehicle comes in and goes out, and realizing a mechanism for automatically changing the weight of the influence after the correction is triggered for multiple times The method has the advantages that the problems that the calculation dimension of the arrival time of the vehicles is single, the influence of passenger flow on the stop time cannot be judged, the influence of vehicle emergency cannot be identified and other factors influence the actual arrival time are solved; the normal work of the electronic stop board is guaranteed by using a low-cost and effective means, and the speed prediction of different road sections is increased; the stop reporting algorithm of the electronic stop board is optimized; vehicle loss handling in special cases: analyzing the influence of traffic lights and passenger flow on arrival of vehicles; establishing a model for influencing the real-time running speed of the vehicle based on multiple factors such as real-time Gps, real-time vehicle-mounted passenger flow, traffic light waiting, vehicle alarming and the like, and performing data training based on different lines and different time periods to obtain accurate prediction of the arrival time of the vehicle at any point from a subsequent station; through the multi-dimensional bus route departure time, the adjacent station interval time, the arrival time and the traffic light waiting time, the connection between the station reporting data and the vehicles is tighter, and the station reporting minutes are more accurate.
Finally, it should be noted that: although the present invention has been described in detail with reference to the foregoing embodiments, it will be apparent to those skilled in the art that modifications may be made to the embodiments or portions thereof without departing from the spirit and scope of the invention.

Claims (10)

1. A method for accurately predicting bus arrival based on big data and dynamic multi-dimension is characterized in that: the method comprises the following steps:
step 1, collecting historical data of buses, and establishing a bus arrival data warehouse of a bus route peak section;
step 2, adopting a device for acquiring the arrival information and the departure information of the vehicles to acquire the arrival information and the departure information of the vehicles, screening unqualified passenger flow data, and inputting the qualified passenger flow data into a data warehouse;
step 3, identifying the use weight, analyzing and calculating the speed weight of the passenger flow to the vehicle on each line, analyzing and calculating the weight of the passenger flow to the station residence time, and taking the weight as an input factor of the measuring and calculating model;
step 4, establishing a bus arrival speed curve chart of each peak section, setting forward movement fluctuation values and backward movement fluctuation values, and comparing the forward movement fluctuation values and the backward movement fluctuation values with the real-time speed;
step 5, cleaning and sorting bus gps deviation correction and vehicle track matching data, associating passenger flow weight, vehicle alarm weight and speed curve graph intervals, using the associated passenger flow weight, vehicle alarm weight and speed curve graph intervals as input factors of a vehicle arrival measuring model of a bus route track, and calculating the time of a vehicle from a first station;
and 6, establishing a training set, and predicting the arrival time of the vehicle to the subsequent station by using the training optimal model.
2. The big data and dynamic multi-dimensional accurate bus arrival prediction method as claimed in claim 1, wherein: the historical data of the public transport vehicle comprises: the bus route departure time, the interval time between adjacent stations, the arrival time and the traffic light waiting time;
the bus departure data acquisition data source is connected with the dispatching system;
the arrival time and the interval time between adjacent stations are obtained by analyzing station entrance and exit data uploaded by a bus-mounted terminal;
the traffic light data acquisition interface is connected with the third-party platform interface.
3. The big data and dynamic multi-dimensional accurate bus arrival prediction method as claimed in claim 1, wherein: the model of the bus arrival data warehouse comprises:
a next station time prediction model for predicting a time when the vehicle is away from the next station;
the inter-station time prediction model is used for predicting the time between the stations in the subsequent interval of the vehicle;
and the N-station time prediction model is used for predicting N-station time after the distance, wherein the N-station time after the distance is the sum of the time prediction of the next station and the time of the subsequent station at intervals.
4. The big data and dynamic multi-dimensional accurate bus arrival prediction method as claimed in claim 1, wherein: the device for acquiring the arrival information and the departure information of the vehicles comprises a vehicle-mounted terminal arranged on the public transport vehicle and used for acquiring the arrival and arrival passenger flow, the departure passenger flow, the first door opening time, the last door closing time and the stop time of the vehicles; the vehicle-mounted terminal is connected with the camera and the sensor, and the sensor identifies the human face behavior to judge the getting-on/off time.
5. The big data and dynamic multi-dimensional accurate bus arrival prediction method as claimed in claim 1, wherein: the screening step of the unqualified passenger flow data comprises the following steps:
deleting data of real-time passenger flow + upper passenger flow-lower passenger flow < 0;
deleting data with the last passenger flow-the next passenger flow exceeding 100;
and deleting data with the last passenger flow or the next passenger flow exceeding 100.
6. The big data and dynamic multi-dimensional accurate bus arrival prediction method as claimed in claim 1, wherein: and step 5, matching the calculated time of the first station with the arrival time of the similar scene of the historical data warehouse, calculating the time of the second station within a controllable range, recording and waiting for arrival in real time and then matching if the input result is not controllable, and recording the abnormity and informing an administrator to change the line or time period influence factor if the time difference between the time and the arrival in real time is greater than a set value.
7. The big data and dynamic multi-dimensional accurate bus arrival prediction method as claimed in claim 1, wherein: the step of analyzing the influence of the passenger flow on the speed of the vehicle on different lines comprises the following steps:
calculating the sectional running speeds of different vehicles on different lines according to the historical gps data of the vehicles;
calculating the passenger flow of different sections of different lines through historical station entrance and exit collection;
and obtaining a linear relation through the speeds of different time segments and the passenger flows of different time segments.
8. The big data and dynamic multi-dimensional accurate bus arrival prediction method as claimed in claim 1, wherein: the input factors of the calculation model comprise:
predicted time to next station T1:
Figure 238707DEST_PATH_IMAGE001
subsequent interval station time prediction T2:
Figure 640870DEST_PATH_IMAGE002
arrival time T = T1+ T2;
Figure 231120DEST_PATH_IMAGE003
wherein the content of the first and second substances,
Figure 958904DEST_PATH_IMAGE004
the mileage between the vehicle and the nearest platform is taken as,
Figure 168693DEST_PATH_IMAGE005
the next station mileage of the vehicle from the nearest station,
Figure 108968DEST_PATH_IMAGE006
the real-time speed of the vehicle is obtained,
Figure 835615DEST_PATH_IMAGE007
in order to influence the factors for the arrival dwell time,
Figure 983569DEST_PATH_IMAGE008
waiting for an impact factor for a traffic light;
Figure 162877DEST_PATH_IMAGE009
the mileage between the nearest platform of the vehicle and the next platform of the vehicle; m is the serial number of the station of the calculation station, m +1 is the serial number of the next station of the calculation station,
Figure 906842DEST_PATH_IMAGE010
the distance between the station and the next station.
9. The big data and dynamic multi-dimensional accurate bus arrival prediction method as claimed in claim 1, wherein: the message format of the bus arrival speed of the peak section is as follows:
00 VVVV _ WW _ X _ Y _ Z, wherein V represents line coding; w represents a service; x represents a date; y represents a peak section; z represents the station code.
10. The big data and dynamic multi-dimensional accurate bus arrival prediction method as claimed in claim 1, wherein: in step 5, the method for cleaning and arranging the public transport gps deviation correction comprises the following steps: calculating the preset time T by the GPS coordinate valueXDynamic mileage C of inner vehicle from starting stationsAnd dynamic mileage C of vehicle from end stationmIf the time T is presetX Increase in the value of (C) and dynamic range CSBy decreasing the value of, or by a predetermined time TX Numerical value of (1) increase dynamic mileage CmIf the numerical value is increased progressively, the data exceeding the preset value of the distance road network offset is eliminated.
CN202210436807.4A 2022-04-25 2022-04-25 Method for accurately predicting bus arrival based on big data and dynamic multidimensional Pending CN114550459A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202210436807.4A CN114550459A (en) 2022-04-25 2022-04-25 Method for accurately predicting bus arrival based on big data and dynamic multidimensional

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202210436807.4A CN114550459A (en) 2022-04-25 2022-04-25 Method for accurately predicting bus arrival based on big data and dynamic multidimensional

Publications (1)

Publication Number Publication Date
CN114550459A true CN114550459A (en) 2022-05-27

Family

ID=81667359

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202210436807.4A Pending CN114550459A (en) 2022-04-25 2022-04-25 Method for accurately predicting bus arrival based on big data and dynamic multidimensional

Country Status (1)

Country Link
CN (1) CN114550459A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115731713A (en) * 2022-11-30 2023-03-03 广东联合电子服务股份有限公司 Method for predicting high-speed exit and time of abnormal vehicle

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115731713A (en) * 2022-11-30 2023-03-03 广东联合电子服务股份有限公司 Method for predicting high-speed exit and time of abnormal vehicle

Similar Documents

Publication Publication Date Title
US20210375128A1 (en) System and method for detection and quantification of irregular traffic congestion
CN114550459A (en) Method for accurately predicting bus arrival based on big data and dynamic multidimensional
CN114662801B (en) Accurate operation management method and system for public transport vehicle
CN109887285B (en) Method and device for determining parking reason
CN115762169A (en) Intelligent control system and method for unmanned driving of sanitation vehicle
CN113838227A (en) Method and system for monitoring abnormal stop of vehicle
CN115050186A (en) Road intelligent rescue management platform based on big data analysis
CN114283590A (en) Traffic flow peak prediction method and device and electronic equipment
CN107480824B (en) Short-time passenger flow prediction system and method for urban rail transit station
CN107025789B (en) Traffic incident detection method and device
CN114627643B (en) Highway accident risk prediction method, device, equipment and medium
CN111932877B (en) Road section traffic abnormal state identification method based on license plate data
CN114639241B (en) Method and system for judging road section interruption state
CN109859458B (en) Vehicle scheduling method and system based on vehicle big data
TWI699662B (en) Detection system and detection method thereof
CN116343484B (en) Traffic accident identification method, terminal and storage medium
CN108182805A (en) A kind of vehicle on highway method for managing and monitoring that biradical modeling is doubted based on letter
CN113085962B (en) Urban rail transit platform safety control method and system
CN115188184B (en) Vehicle speed limit processing method, device and apparatus
CN112382084B (en) Comprehensive emergency management platform architecture for cloud computing
CN110619745B (en) Data processing method and device for preventing traffic jam
CN114580787B (en) Passenger flow prediction method and device based on bus subway connection station
CN111932893B (en) Method for realizing road section state studying and judging processing based on signal and electric police data fusion technology
CN117953715A (en) Smart city traffic management system based on big data analysis
CN117523819A (en) Vehicle delay prediction method and system based on data analysis

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
RJ01 Rejection of invention patent application after publication
RJ01 Rejection of invention patent application after publication

Application publication date: 20220527