KR20170003769U - An Error Correction Model for Bus Arrival Information Prediction System Using Machine Learning - Google Patents

An Error Correction Model for Bus Arrival Information Prediction System Using Machine Learning Download PDF

Info

Publication number
KR20170003769U
KR20170003769U KR2020170005328U KR20170005328U KR20170003769U KR 20170003769 U KR20170003769 U KR 20170003769U KR 2020170005328 U KR2020170005328 U KR 2020170005328U KR 20170005328 U KR20170005328 U KR 20170005328U KR 20170003769 U KR20170003769 U KR 20170003769U
Authority
KR
South Korea
Prior art keywords
bus
information
time
arrival
variables
Prior art date
Application number
KR2020170005328U
Other languages
Korean (ko)
Inventor
김재식
Original Assignee
김재식
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 김재식 filed Critical 김재식
Priority to KR2020170005328U priority Critical patent/KR20170003769U/en
Publication of KR20170003769U publication Critical patent/KR20170003769U/en

Links

Images

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06QINFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q50/00Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
    • G06Q50/40Business processes related to the transportation industry
    • G06Q50/30
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06NCOMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00Computing arrangements based on biological models
    • G06N3/02Neural networks
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06QINFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q50/00Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
    • G06Q50/10Services
    • G06Q50/26Government or public services
    • GPHYSICS
    • G08SIGNALLING
    • G08GTRAFFIC CONTROL SYSTEMS
    • G08G1/00Traffic control systems for road vehicles
    • G08G1/123Traffic control systems for road vehicles indicating the position of vehicles, e.g. scheduled vehicles; Managing passenger vehicles circulating according to a fixed timetable, e.g. buses, trains, trams

Landscapes

  • Engineering & Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Tourism & Hospitality (AREA)
  • General Health & Medical Sciences (AREA)
  • Health & Medical Sciences (AREA)
  • Strategic Management (AREA)
  • Marketing (AREA)
  • Human Resources & Organizations (AREA)
  • Economics (AREA)
  • Primary Health Care (AREA)
  • General Business, Economics & Management (AREA)
  • Biomedical Technology (AREA)
  • Data Mining & Analysis (AREA)
  • General Engineering & Computer Science (AREA)
  • Molecular Biology (AREA)
  • Mathematical Physics (AREA)
  • Software Systems (AREA)
  • Evolutionary Computation (AREA)
  • Development Economics (AREA)
  • Educational Administration (AREA)
  • Computing Systems (AREA)
  • Computational Linguistics (AREA)
  • Biophysics (AREA)
  • Artificial Intelligence (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Remote Sensing (AREA)
  • Radar, Positioning & Navigation (AREA)
  • Operations Research (AREA)
  • Traffic Control Systems (AREA)

Abstract

현재의 버스 정보 시스템에 구축되어 있는 버스 도착 예상시간 산출 모형은 통행시간에 영향을 미치는 교통과 환경 상황이 반영되어 있지 않아 상황 변화에 따라 많은 오차를 발생시키고 있음. 선행 연구 사례를 통해 다양한 교통 변수를 활용하면 더욱 정교하고 개선된 정보를 제공할 수 있다는 점을 바탕으로, 기존의 산출 모형들의 장점에 더불어 획득 가능한 다양한 교통과 환경 변수들을 통하여 상황에 맞게 유동적으로 예측 시간 정보를 제공할 수 있는 예상시간 추정 시스템을 제안하고자 함.The Bus Estimated Time Estimation Model, which is built on the current bus information system, does not reflect the traffic and environment conditions affecting the travel time, and thus causes a lot of error according to the situation change. Based on the fact that various traffic variables can provide more sophisticated and improved information through the precedent research case, it is possible to use the various traffic and environment variables that can be obtained along with the advantages of existing calculation models, We propose a prediction time estimation system that can provide information.

Figure utm00001
Figure utm00001

Description

머신러닝을 이용한 버스 도착 정보 예측 시스템의 오차 개선 모델 {An Error Correction Model for Bus Arrival Information Prediction System Using Machine Learning}An Error Correction Model for Bus Arrival Information Prediction System Using Machine Learning.

현재의 버스 정보 시스템에 구축되어 있는 버스 도착 예상시간 산출 모형은 통행시간에 영향을 미치는 교통과 환경 상황이 반영되어 있지 않아 상황 변화에 따라 많은 오차를 발생시키고 있음. 선행 연구 사례를 통해 다양한 교통 변수를 활용하면 더욱 정교하고 개선된 정보를 제공할 수 있다는 점을 바탕으로, 기존의 산출 모형들의 장점에 더불어 획득 가능한 다양한 교통과 환경 변수들을 신경망 머신러닝을 통하여 상황에 맞게 유동적으로 예측 시간 정보를 제공할 수 있는 예상시간 추정 시스템을 제안하고자 함.The Bus Estimated Time Estimation Model, which is built on the current bus information system, does not reflect the traffic and environment conditions affecting the travel time, and thus causes a lot of error according to the situation change. Based on the fact that various traffic variables can provide more sophisticated and improved information through previous research examples, various traffic and environment variables that can be obtained along with the advantages of existing calculation models can be obtained through neural network machine learning We propose a prediction time estimation system that can provide predictive time information in a flexible way.

국내에서는 과거의 시계열 자료를 활용하는 단순/(가중)이동 평균법이 가장 많이 운용중이다. 이는관측 값에 보다 큰 가중치를 부여하여 현재의 소통상황을 반영한 예측값을산출할 수 있도록 하는 장점이 있으나 이러한 시스템은 교통량이 급변하거나 소요시간이 크게 변동할 경우, 예를 들어 출,퇴근 시간이나 매우 한적한 경우 오차량 급증하는 현상이 발생한다. 또한 비정상적으로 운행된 특정 버스의 소요시간이 반영되면 제공정보의 오류가 발생하며 오차량이 크게 상승할 경우 지연오차와 조기도착오차의 반복 패턴 형성한다.In Korea, the simple / (weighted) moving average method, which utilizes past time series data, is the most widely used. This system has the advantage of giving a larger weight to the observed values, so that it can calculate the predicted value reflecting the current traffic situation. However, such a system has a problem in that when the traffic volume changes suddenly or the time required is greatly changed, In the case of cold, the phenomenon that the vehicle increases suddenly occurs. Also, if the time required for the abnormal bus is reflected, the supplied information is erroneous, and if the erroneous vehicle is greatly elevated, a repeated pattern of delay error and early arrival error is formed.

운용 중인시스템의 예상시간 산출법은 가중이동평균법으로 최근 운행한 버스들의 이동 구간에 대한 운행 소요시간의 평균값을 예상시간으로 제공해주는데, 이 경우 교통량이 급변하여 소요시간이 크게 변동할 경우에는오차량이급증하고,지연 오차와조기 도착 오차가반복되는 악순환을 만들게 된다.The method of calculating the expected time of the operating system provides the average value of the time required for the travel of the buses that have recently been operated by the weighted moving average method as the estimated time. In this case, when the amount of traffic greatly changes, And a vicious cycle in which the delay error and the early arrival error are repeated.

이에 따라 시계열 자료의최근 이력을 통한 산출방식과 달리, 운행시간에 직/간접적으로 영향을 미치는 날씨의 변화, 도로의 상태, 대중교통 이용자 수의 변화 등의 교통 환경 상황을 중심으로 비슷한 환경에서 운행된 과거 이력의 운행 소요시간을 평균대상 값으로활용하여 시시각각 변화하는 교통 상태에 실시간으로 대응된결괏값을제공하여 오차 보정과 낮은 오차분산으로 시스템 신뢰도를향상시킬수 있다. Therefore, unlike the method of calculation using the recent history of time series data, it operates in a similar environment centering on the traffic environment conditions such as change of weather, change of the road condition and number of public transportation users, which directly / indirectly affect the running time By using the time required for running the past history as the average target value, it is possible to improve the system reliability by correcting the error and providing low error dispersion by providing the correspondence value corresponding to the changing traffic state in real time.

기존 플렛폼에 의한 시간 버스 도착 예정 시간을 반영 하는 대신 시시각각 변화하는 교통/환경 상태에 실시간으로 대응된 결과값을 제공할 수 있으며 더불어 축적 데이터의 표본 산출을 통한 오차 보정과 낮은 오차분산으로 시스템 신뢰도 향상에 기여한다. 또한 신규 하드웨어의 증설없이 소프트웨어의 개선만으로 오차개선 효과를 나타낼 수 있으며 이에 따른 정시성 확보를 통한 이용자의 불편함 감소에 기여, 실추된 버스의 위상 제고를 기대한다.Instead of reflecting the estimated arrival time of the bus by the existing platform, it can provide the result value corresponding to the changing traffic / environment condition in real time. In addition, it can improve the system reliability by correcting the error through the sampling of the accumulated data and low error dispersion. . In addition, it is possible to improve the error by only improving the software without adding new hardware, and contributing to the reduction of user inconvenience by securing timeliness and expecting the improvement of the lost bus.

또한 상황에 따른 구간별 교통시간 제공을 통해 효율적인 대중교통 자원을 선택함으로써 전체적인 경제성을 향상시키며 소프트웨어 개선만으로 오차보정이 가능하므로 국내 시스템 기능개선 및 성능 개량(7년단위)시 반영이 가능하다는 장점이있으며 해외 업체 대비 정확/우수한 정보 제공 가능을 도모하여 해외 경쟁력 확보가 가능하다In addition, it can improve the overall economic efficiency by selecting efficient public transportation resources by providing traffic time for each section according to the situation, and it is possible to correct the errors by only software improvement, so that it is possible to improve the domestic system functions and improve the performance It is possible to secure overseas competitiveness by providing accurate / excellent information compared to overseas companies.

[도 1]은 차량, 정류장, 국가교통정보센터, 기상청, 경찰청, 지방자치단체 등과 같은 시설이나 건물에서부터 운행 상태의 정보, 현재 위치, 정류장 통과 정보, 승/하차 소요시간, 교통량, 차로수, 신호등 수, 시상 상태, 강수량, 사고정보, 도로통제정보등의 데이터등을 받아들이는 과정을 표시하며 이러한 결과로 축적된 데이터는 학습과 결과 도출을 위해 사용되어진다.FIG. 1 is a diagram showing the information on the state of operation, current location, stop passage information, time required for getting on and off, traffic volume, number of lanes, number of lanes, and the like, from facilities and buildings such as a vehicle, a bus stop, a national traffic information center, The number of traffic lights, the number of traffic lights, the number of traffic lights, precipitation, accident information, and road control information.

버스의 구간별 통과 정보 저장 시 구간별 평균 운행속도, 해당 도로 교통량, 기상 상태 등 운행시간에 영향을 끼치는 변수들을 함께 축적시켜 두었다가, 추후 버스 도착정보 산출 시에 누적된 이력에서 산출하고자 하는 노선의 실시간 교통 환경 상황이 비슷한 과거의 시점을 변수들을 통해 조회한 후 조회된 시점에서의 운행시간과 승/하차소요시간 등을 가공하여 예상시간을 산출해 정류장단말기(BIT)와 API를 통해 이용자에게 제공토록 한다.It is necessary to accumulate variables influencing the running time such as the average running speed, the road traffic volume, the weather condition, etc. during the storage of the pass information of the bus by the interval, It estimates the estimated time by processing the operating time and the time required to get off and on from the viewpoint after viewing the past point of time similar to the real-time traffic environment situation through the variables and providing it to the user through the station terminal (BIT) and the API I will.

Claims (5)

버스 운행 관련 변수를 획득하는 획득부,

상기 정보 획득부에 연결되어 있으며 도착 예정 정보를 관리 서버로 전송하고, 상기 관리 서버로 부터 데이터를 인공지능 시스템을 이용하여 데이터를 가공, 수신하는 동작 제어부
를 포함하는 실시간 버스 정보 시스템 도착 예측 시스템의 오차 개선 모델 시스템
An acquisition unit for acquiring a bus operation related parameter,

An operation control unit connected to the information obtaining unit and transmitting arrival planning information to the management server and processing data from the management server using the artificial intelligence system,
Error-correcting model system of real-time bus information system arrival forecasting system
제 1항에서,
상기 동작 제어부는 기존에 상용되고 있는 버스 도착 예정 산출 모형을 바탕으로 개선된 알고리즘으로 혼합모델을 사용
The method of claim 1,
The operation control unit uses the mixed model as an improved algorithm based on the existing bus arrival forecast calculation model which is commonly used
제 1항에서,
버스 도착 2, 3전 정류장 외 버스 도착 예정 정보는 획득부를 통한 데이터를 통한 신경망 알고리즘을 통한 도착 예정 정보를 반영 이 후 2, 3전 내로 버스가 들어 오게 되면 기존의 모형인 가중이동평균법을 사용.
The method of claim 1,
Bus Arrival 2 and 3 The bus arrival information outside the bus stop reflects arrival information through the neural network algorithm using data from the acquisition unit. When the bus arrives within 2 or 3 hours, the existing model, the weighted moving average method, is used.
제 1항에서,
상기 획득부는 기상청, 경찰청, 국가교통정보센터, 각 기관처로 부터 받아들이는 운행 관련 직/간접 변수를 받아들이는 서버.
The method of claim 1,
The acquisition unit is a server that receives operational and related direct / indirect variables received from the Korea Meteorological Administration, the National Police Agency, the National Traffic Information Center, and the respective agencies.
제 4항에서 ,
버스의 구간별 통과 정보 저장 시 구간별 평균 운행속도, 해당 도로 교통량, 기상 상태 등 운행시간에 영향을 끼치는 변수들을 함께 축적시켜 두었다가, 동작 제어부를 통하여 추후 버스 도착정보 산출 시에 누적된 이력에서 산출하고자 하는 노선의 실시간 교통 환경 상황이 비슷한 과거의 시점을 변수들을 통해 조회한 후 조회된 시점에서의 운행시간과 승/하차소요시간 등을 가공하여 예상시간을 산출해 정류장단말기(BIT)와 API를 통해 이용자에게 제공토록 한다.
5. The method of claim 4,
The variables affecting the running time such as the average running speed, the road traffic volume, the weather condition, and the like during the storage of the pass information for each section of the bus are stored together, and then calculated from the cumulative history at the time of calculating the bus arrival information (BIT) and the API (BIT) by calculating the estimated time by processing the operating time and the time required for getting off and on from the viewpoint after viewing the similar past point of time through the variables, To the user.
KR2020170005328U 2017-10-14 2017-10-14 An Error Correction Model for Bus Arrival Information Prediction System Using Machine Learning KR20170003769U (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
KR2020170005328U KR20170003769U (en) 2017-10-14 2017-10-14 An Error Correction Model for Bus Arrival Information Prediction System Using Machine Learning

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
KR2020170005328U KR20170003769U (en) 2017-10-14 2017-10-14 An Error Correction Model for Bus Arrival Information Prediction System Using Machine Learning

Publications (1)

Publication Number Publication Date
KR20170003769U true KR20170003769U (en) 2017-11-01

Family

ID=60408279

Family Applications (1)

Application Number Title Priority Date Filing Date
KR2020170005328U KR20170003769U (en) 2017-10-14 2017-10-14 An Error Correction Model for Bus Arrival Information Prediction System Using Machine Learning

Country Status (1)

Country Link
KR (1) KR20170003769U (en)

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109920248A (en) * 2019-03-05 2019-06-21 南通大学 A kind of public transport arrival time prediction technique based on GRU neural network
KR20190093785A (en) 2018-01-16 2019-08-12 국민대학교산학협력단 Method for adaptive machine learning in iot
CN110459056A (en) * 2019-08-26 2019-11-15 南通大学 A kind of public transport arrival time prediction technique based on LSTM neural network
KR102397198B1 (en) * 2021-09-03 2022-05-13 한국과학기술정보연구원(Kisti) Bus operation time prediction device and the operation method

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR20080062998A (en) * 2006-12-29 2008-07-03 포스데이타 주식회사 Traffic information guiding system and method
KR20160141114A (en) * 2015-05-28 2016-12-08 한국전자공업협동조합 Bus information system and method

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR20080062998A (en) * 2006-12-29 2008-07-03 포스데이타 주식회사 Traffic information guiding system and method
KR20160141114A (en) * 2015-05-28 2016-12-08 한국전자공업협동조합 Bus information system and method

Non-Patent Citations (3)

* Cited by examiner, † Cited by third party
Title
남여민 외 2인_기계학습을 이용한 버스도착예정시간 예측_한국정보확학회 2016년 동계학술발표회 논문집_pp.708~710 (2016.12) *
이주영 외 3인, BIS 자료를 이용한 중장기 버스 통행시간 예측, 대한교통학회지 제35권 제4호, pp.348~359 (2017.08) *
이주영 외 3인_BIS 자료를 이용한 중장기 버스 통행시간 예측_대한교통학회지 제35권 제4호_pp.348~359 (2017.08) *

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR20190093785A (en) 2018-01-16 2019-08-12 국민대학교산학협력단 Method for adaptive machine learning in iot
CN109920248A (en) * 2019-03-05 2019-06-21 南通大学 A kind of public transport arrival time prediction technique based on GRU neural network
CN109920248B (en) * 2019-03-05 2021-09-17 南通大学 Bus arrival time prediction method based on GRU neural network
CN110459056A (en) * 2019-08-26 2019-11-15 南通大学 A kind of public transport arrival time prediction technique based on LSTM neural network
KR102397198B1 (en) * 2021-09-03 2022-05-13 한국과학기술정보연구원(Kisti) Bus operation time prediction device and the operation method

Similar Documents

Publication Publication Date Title
KR20170003769U (en) An Error Correction Model for Bus Arrival Information Prediction System Using Machine Learning
CN103310651B (en) A kind of public transport based on real-time road condition information is arrived at a station Forecasting Methodology
GB2599765A (en) Vehicle traffic flow prediction method with missing data
CN102521989B (en) Dynamic-data-driven highway-exit flow-quantity predicting method
Potter et al. Potential benefits of a dedicated probabilistic rapid ramp event forecast tool
JP6065531B2 (en) Energy consumption profiling
CN109724611A (en) Paths planning method, device, electronic equipment and storage medium
CN108399741A (en) A kind of intersection flow estimation method based on real-time vehicle track data
CN102316496A (en) Data merging method based on Kalman filtering in wireless sensor network
JP2019204458A (en) Electric power demand prediction system, learning device, and electric power demand prediction method
CN111882867A (en) City wisdom traffic early warning system based on thing networking
CN105096642A (en) Real-time bus arrival time prediction method in consideration of GPS data delay effect
CN104318327A (en) Predictive parsing method for track of vehicle
CN110738855A (en) Road traffic flow condition prediction method for data sparse time periods
CN109784559B (en) Method for calculating cumulative damage fault probability of transmission tower under typhoon path prediction information
CN109859515A (en) GPS positioning compensation data method and electronic equipment in public transit system
CN109064742A (en) A kind of adaptive public transport arrival time prediction technique based on SVM
CN109697304A (en) A kind of construction method of refrigeration unit multi-sensor data prediction model
CN115829183A (en) Cold-chain logistics path planning method, device, equipment and readable storage medium
CN112686417A (en) Subway large passenger flow prediction method and system and electronic equipment
Alsumaiei Short-term forecasting of monthly water consumption in hyper-arid climate using recurrent neural networks
CN107479112A (en) Thick fog Forecasting Methodology, apparatus and system
CN115619045B (en) Estuary tidal range forecasting method and system
CN112052339A (en) Target track prediction method, system and storage medium based on knowledge graph
CN108682152B (en) Bus abnormal data restoration method based on preceding bus data

Legal Events

Date Code Title Description
A201 Request for examination
E902 Notification of reason for refusal
E902 Notification of reason for refusal
E601 Decision to refuse application
E601 Decision to refuse application