[논문]중심합성계획법 기반 이중유전자알고리즘을 활용한 차량추종모형 정산방법론 개발

배범준; 임현섭; 소재현

doi:10.12815/kits.2019.18.2.29

중심합성계획법 기반 이중유전자알고리즘을 활용한 차량추종모형 정산방법론 개발
Calibration of Car-Following Models Using a Dual Genetic Algorithm with Central Composite Design 원문보기

韓國ITS學會論文誌 = The journal of the Korea Institute of Intelligent Transportation Systems, v.18 no.2, 2019년, pp.29 - 43

배범준 (한국교통연구원 민자도로관리연구센터) , 임현섭 , 소재현 (한국교통연구원 스마트시티교통연구팀)

초록
AI-Helper

미시적 교통류 모형의 정산은 시뮬레이션 분석에 있어 매우 중요한 요소이다. 유전자 알고리즘은 교통류 모형의 정산에 널리 활용되어 왔으며, 일반적으로 이러한 최적화 문제에 있어 높은 효율성을 보이는 것으로 알려져 있다. 하지만 제한된 시간내에 신속한 의사결정을 위한 시뮬레이션 분석에 있어 유전자알고리즘의 모형 정산속도는 여전히 느리다. 이에 본 연구에서는 정산 효율 향상을 위해 중심합성계획법 기반의 이중유전자알고리즘을 활용한 차량추종모형 정산방법론을 개발하였다. 개발된 정산 방법론에서는 실험계획법 중 하나인 중심합성계획법과 유전자알고리즘을 결합하여 준최적해를 찾고, 이를 다시 유전자알고리즘의 초기 값으로 하여 모형 파라미터의 최적해를 찾는다. 개발된 방법을 활용하여 Gipps의 차량추종모형을 정산하였다. 선행연구에서 사용된 단일 유전자알고리즘을 활용한 방법과 비교한 결과, 본 연구에서 개발한 방법이 더 짧은 시간내에 최적해를 찾는 것으로 확인되었다. 개발된 방법론은 유전자알고리즘을 사용하는 다양한 교통분석에 활용될 수 있을 것으로 기대된다.

Abstract ▼ AI-Helper

The calibration of microscopic traffic simulation models has received much attention in the simulation field. Although no standard has been established for it, a genetic algorithm (GA) has been widely employed in recent literature because of its high efficiency to find solutions in such optimization problems. However, the performance still falls short in simulation analyses to support fast decision making. This paper proposes a new calibration procedure using a dual GA and central composite design (CCD) in order to improve the efficiency. The calibration exercise goes through three major sequential steps: (1) experimental design using CCD for a quadratic response surface model (RSM) estimation, (2) 1st GA procedure using the RSM with CCD to find a near-optimal initial population for a next step, and (3) 2nd GA procedure to find a final solution. The proposed method was applied in calibrating the Gipps car-following model with respect to maximizing the likelihood of a spacing distribution between a lead and following vehicle. In order to evaluate the performance of the proposed method, a conventional calibration approach using a single GA was compared under both simulated and real vehicle trajectory data. It was found that the proposed approach enhances the optimization speed by starting to search from an initial population that is closer to the optimum than that of the other approach. This result implies the proposed approach has benefits for a large-scale traffic network simulation analysis. This method can be extended to other optimization tasks using GA in transportation studies.

주제어

AI 본문요약
AI-Helper

* AI 자동 식별 결과로 적합하지 않은 문장이 있을 수 있으니, 이용에 유의하시기 바랍니다.

문제 정의

Since the task of simulation in this study assumes that no disaggregated level attribute is available, the simulation model should be calibrated in a way to represent similar distribution of the aggregated attributes that can be observed in collected data. In this sense, the objective of this optimization task is to maximize the likelihood of having simulation results same as values in observed data. For each time point, a probability density of having a predicted value from the simulation using the KDE from the observations can be calculated.
Many other options in the proposed method should be investigated for future studies in order to enhance the improvement in this paper. Running GA twice in the proposed algorithm brings additional complexity in computation, compared to running a single GA.

제안 방법

A genetic algorithm (GA) is a search algorithm that emulates biological evolutionary theories. A GA searches solutions at multiple locations in a parameter space based on three probabilistic rules including selection, crossover, and mutation over generations (or iterations). Thus, the main advantage of GA is that it has a low likelihood of getting stuck in local optima while providing high computational efficiency(Kim et al.
It was assumed that those parameter values are correct for this study. Since the purpose of this study is to propose a new approach for model calibration and verify its characteristics, this assumption allows to avoid judging the accuracy of the Gipps.
To get the real vehicle trajectory data, the latitude/longitude data and the corresponding speed of a vehicle were collected every second using a GPS data logger while driving. The data collection was carried out during off-peak hours of a weekday on an arterial in an urban area. Therefore, it includes comprehensive driving behaviors in terms of speed and acceleration.
Using the CCD, GA can start to search from the area that is closer to the optimum so that it can skip initial iterations of a simulation run. The higher efficiency of the proposed method was identified by comparing with the single GA approach which is the convention in traffic simulation area with the simulated and real vehicle trajectory data.
1]. The major characteristics of this approach is to use CCD for estimating a quadratic response model and to employ dual GA on the basis of the response modeling result.
A central composite design (CCD) is the most frequently utilized second-order design for RSM. This design consists of three types of experimental runs: a two-factor factorial design, center points and axial points. A fractional factorial is generally used in CCD for five or more factors(Mee, 2009).
This paper proposes a hybrid calibration procedure in which response surface methodology (RSM) with central composite design (CCD) is employed with GA to improve efficiency of the calibration process. Using field-collected vehicle trajectory data, the procedure is applied to calibrate the parameters of Gipps car-following model in order to fit spacing distribution between lead and following vehicles.
This paper proposes a hybrid calibration procedure in which response surface methodology (RSM) with central composite design (CCD) is employed with GA to improve efficiency of the calibration process. Using field-collected vehicle trajectory data, the procedure is applied to calibrate the parameters of Gipps car-following model in order to fit spacing distribution between lead and following vehicles.

대상 데이터

The United States Department of Transportation (US DOT) Federal Highway Administration (FHWA) launched the NGSIM program in 2005 to develop microscopic traffic behaviour models. As a part of this program, the US 101 trajectory dataset(FHWA,2016) was collected on southbound US 101, the Hollywood Freeway in California. Every individual vehicle’s trajectory data were collected on the 640-meter (2,100-feet) length of study area during the morning peak hours.

이론/모형

The GA has been used widely in calibration of simulation models, to solve the problem using an automated procedure. In this study, a GA solver of the optimization toolbox in Matlab was used to find the best solution for the Gipps model.

LOADING...

내보내기 구분	파일저장 인쇄 메일전송
구성항목	기본정보 상세정보 관리번호, 논문명, 저널/프로시딩명, 저자 , 발행년, 권, 호, 시작페이지, 끝페이지, 발행기관 관리번호, 논문명, 대등논문명, 저자 , 저널/프로시딩명, 발행기관, 발행년, 발행언어, 권, 호, 시작페이지, 끝페이지, ISBN, ISSN, 주제분야, 키워드, 초록(한글), 초록(영문), 저자(소속기관)
저장형식	Text(ASCII format) Excel format RefWorks Direct Export RIS format (for Reference Manager, ProCite, EndNote), Scholar's Aids, Mendeley
메일정보	받는사람 (필수) @ 보내는사람 (선택) @ 제목 내용 KISTI 검색결과 이메일 서비스
안내	총 건의 자료가 검색되었습니다. 다운받으실 자료의 인덱스를 입력하세요. (1-10,000) 검색결과의 순서대로 최대 10,000건 까지 다운로드가 가능합니다. 데이타가 많을 경우 속도가 느려질 수 있습니다.(최대 2~3분 소요) 다운로드 파일은 UTF-8 형태로 저장됩니다. 파일의 내용이 제대로 보이지 않을실 때는 웹브라우저 상단의 보기 -> 인코딩 -> 자동선택 여부를 확인하십시오. ~ Text(ASCII format) Excel format

연합인증

중심합성계획법 기반 이중유전자알고리즘을 활용한 차량추종모형 정산방법론 개발
Calibration of Car-Following Models Using a Dual Genetic Algorithm with Central Composite Design 원문보기

초록
AI-Helper

Abstract ▼ AI-Helper

주제어

AI 본문요약
AI-Helper

문제 정의

제안 방법

대상 데이터

이론/모형

이 논문을 인용한 문헌

연구과제 타임라인

관련 콘텐츠

원문 보기

원문 URL 링크

오픈액세스(OA) 유형

연관된 기능

이 논문과 함께 이용한 콘텐츠

AI-Helper ※ AI-Helper는 오픈소스 모델을 사용합니다.

선택된 텍스트

연합인증

중심합성계획법 기반 이중유전자알고리즘을 활용한 차량추종모형 정산방법론 개발 Calibration of Car-Following Models Using a Dual Genetic Algorithm with Central Composite Design 원문보기

초록 용어보기논문에서 용어와 풀이말을 자동 추출한 결과로, 시범 서비스 중입니다. AI-Helper

Abstract ▼ AI-Helper

주제어

AI 본문요약 엑셀 다운로드 AI-Helper

문제 정의

제안 방법

대상 데이터

이론/모형

이 논문을 인용한 문헌

연구과제 타임라인

전체(0) 논문(0) 특허(0) 보고서(0)

전체(0) 논문(0) 특허(0) 보고서(0)

관련 콘텐츠

원문 보기

원문 URL 링크

오픈액세스(OA) 유형

연관된 기능

이 논문과 함께 이용한 콘텐츠

AI-Helper ※ AI-Helper는 오픈소스 모델을 사용합니다.

선택된 텍스트

중심합성계획법 기반 이중유전자알고리즘을 활용한 차량추종모형 정산방법론 개발
Calibration of Car-Following Models Using a Dual Genetic Algorithm with Central Composite Design 원문보기

초록
AI-Helper

AI 본문요약
AI-Helper