서브메뉴
검색
Fundamental Diagram for Mixed Traffic Flow: Stochastic and Dynamic Properties
Fundamental Diagram for Mixed Traffic Flow: Stochastic and Dynamic Properties
상세정보
- 자료유형
- 학위논문 서양
- 최종처리일시
- 20260202104710
- ISBN
- 9798286456482
- DDC
- 385
- 저자명
- Jiang, Jiwan.
- 서명/저자
- Fundamental Diagram for Mixed Traffic Flow: Stochastic and Dynamic Properties
- 발행사항
- [Sl] : The University of Wisconsin - Madison, 2025
- 발행사항
- Ann Arbor : ProQuest Dissertations & Theses, 2025
- 형태사항
- 132 p
- 주기사항
- Source: Dissertations Abstracts International, Volume: 87-01, Section: B.
- 주기사항
- Advisor: Ahn, Soyoung Sue.
- 학위논문주기
- Thesis (Ph.D.)--The University of Wisconsin - Madison, 2025.
- 초록/해제
- 요약With the increasing integration of automated vehicles (AVs) into existing traffic systems predominantly composed of human-driven vehicles (HDVs), understanding mixed traffic dynamics becomes critically important for efficient and stable transportation management. This dissertation aims to elucidate the stochastic and dynamic properties of traffic flow, mixed with HDVs and AVs. To achieve this, the study: (1) investigates the stochastic car-following features inherent to both HDVs and AVs; (2) analytically derives a stochastic dynamic fundamental diagram (FD) for AV traffic based on a well-known car following control model; and (3) extends the above FD framework to mixed traffic by incorporating the stochastic and nonlinear properties characteristic of HDVs.Car following (CF) models are crucial for understanding traffic dynamics; however, human drivers' CF behaviors are unpredictable and nonlinear. Identifying the best CF model remains challenging despite extensive research, compounded by the introduction of automated vehicles with proprietary controllers. To address this, the present study introduces a stochastic learning framework based on Approximate Bayesian Computation (ABC), which systematically integrates multiple CF models into a hybrid representation. This approach evaluates and combines models according to their likelihood of accurately describing observed driving behaviors, thus enhancing interpretability and predictive accuracy. Evaluations conducted using two distinct datasets demonstrate that this hybrid model framework significantly improves trajectory reproduction for both HDVs and AVs compared to traditional single-model approaches.The fundamental diagram (FD) of traffic flow describes intrinsic relationships among flow, density, and speed, serving as a foundational tool for traffic analysis. Advancing beyond static FD representations, this study analytically formulates a dynamic FD by deriving it directly from vehicle-level CF (control) laws, thus capturing complex dynamic phenomena such as traffic hysteresis. This analytical derivation leverages a frequency-domain representation of vehicle kinematics coupled with continuum approximations of macroscopic traffic variables. The proposed formulation is sufficiently general to accommodate any analytically describable CF law applicable to either HDVs or AVs. Numerical experiments shed light on how the choice of density-flow measurement regions and CF parameters influence dynamic FD properties within AV platoons.Building upon these developments, this research further extends the FD framework to mixed traffic environments, explicitly incorporating the stochastic and nonlinear CF characteristics of HDVs. Utilizing describing function analysis (DFA), approximate linear transfer functions are derived for nonlinear HDV CF models. Subsequently, a sequence-based stochastic dynamic FD is established for mixed platoons, enabling quantification of density and flow under varying vehicle sequencing configurations and AV penetration rates. Given the complexity of the stochastic interactions and the absence of closed-form solutions, Monte Carlo simulations are employed to characterize the resulting FD properties comprehensively. Findings from these simulations indicate that HDVs introduce greater randomness and variability, whereas AVs produce more structured, though still oscillatory, dynamic patterns. This underscores the necessity of carefully managing AV integration strategies to mitigate unintended amplification of traffic oscillations.In conclusion, this dissertation establishes a systematic and integrated framework for modeling and understanding mixed traffic dynamics, effectively capturing both the stochasticity inherent in human driver behaviors and the nonlinear dynamics introduced by diverse AV control strategies. The outcomes of this research provide critical insights and methodological contributions that support the development of advanced traffic modeling tools and simulation approaches. Ultimately, these findings contribute to a comprehensive understanding of traffic dynamics in mixed-vehicle contexts, informing future traffic management strategies and supporting the broader transition towards increasingly automated transportation systems.
- 일반주제명
- Transportation
- 일반주제명
- Engineering
- 일반주제명
- Automotive engineering
- 키워드
- Car following
- 키워드
- Mixed traffic
- 기타저자
- The University of Wisconsin - Madison Civil & Environmental Engr
- 기본자료저록
- Dissertations Abstracts International. 87-01B.
- 전자적 위치 및 접속
- 로그인 후 원문을 볼 수 있습니다.
MARC
008260126s2025 us c eng d■001000017358499
■00520260202104710
■006m o d
■007cr#unu||||||||
■020 ▼a9798286456482
■035 ▼a(MiAaPQ)AAI32118587
■040 ▼aMiAaPQ▼cMiAaPQ
■0820 ▼a385
■1001 ▼aJiang, Jiwan.
■24510▼aFundamental Diagram for Mixed Traffic Flow: Stochastic and Dynamic Properties
■260 ▼a[Sl]▼bThe University of Wisconsin - Madison▼c2025
■260 1▼aAnn Arbor▼bProQuest Dissertations & Theses▼c2025
■300 ▼a132 p
■500 ▼aSource: Dissertations Abstracts International, Volume: 87-01, Section: B.
■500 ▼aAdvisor: Ahn, Soyoung Sue.
■5021 ▼aThesis (Ph.D.)--The University of Wisconsin - Madison, 2025.
■520 ▼aWith the increasing integration of automated vehicles (AVs) into existing traffic systems predominantly composed of human-driven vehicles (HDVs), understanding mixed traffic dynamics becomes critically important for efficient and stable transportation management. This dissertation aims to elucidate the stochastic and dynamic properties of traffic flow, mixed with HDVs and AVs. To achieve this, the study: (1) investigates the stochastic car-following features inherent to both HDVs and AVs; (2) analytically derives a stochastic dynamic fundamental diagram (FD) for AV traffic based on a well-known car following control model; and (3) extends the above FD framework to mixed traffic by incorporating the stochastic and nonlinear properties characteristic of HDVs.Car following (CF) models are crucial for understanding traffic dynamics; however, human drivers' CF behaviors are unpredictable and nonlinear. Identifying the best CF model remains challenging despite extensive research, compounded by the introduction of automated vehicles with proprietary controllers. To address this, the present study introduces a stochastic learning framework based on Approximate Bayesian Computation (ABC), which systematically integrates multiple CF models into a hybrid representation. This approach evaluates and combines models according to their likelihood of accurately describing observed driving behaviors, thus enhancing interpretability and predictive accuracy. Evaluations conducted using two distinct datasets demonstrate that this hybrid model framework significantly improves trajectory reproduction for both HDVs and AVs compared to traditional single-model approaches.The fundamental diagram (FD) of traffic flow describes intrinsic relationships among flow, density, and speed, serving as a foundational tool for traffic analysis. Advancing beyond static FD representations, this study analytically formulates a dynamic FD by deriving it directly from vehicle-level CF (control) laws, thus capturing complex dynamic phenomena such as traffic hysteresis. This analytical derivation leverages a frequency-domain representation of vehicle kinematics coupled with continuum approximations of macroscopic traffic variables. The proposed formulation is sufficiently general to accommodate any analytically describable CF law applicable to either HDVs or AVs. Numerical experiments shed light on how the choice of density-flow measurement regions and CF parameters influence dynamic FD properties within AV platoons.Building upon these developments, this research further extends the FD framework to mixed traffic environments, explicitly incorporating the stochastic and nonlinear CF characteristics of HDVs. Utilizing describing function analysis (DFA), approximate linear transfer functions are derived for nonlinear HDV CF models. Subsequently, a sequence-based stochastic dynamic FD is established for mixed platoons, enabling quantification of density and flow under varying vehicle sequencing configurations and AV penetration rates. Given the complexity of the stochastic interactions and the absence of closed-form solutions, Monte Carlo simulations are employed to characterize the resulting FD properties comprehensively. Findings from these simulations indicate that HDVs introduce greater randomness and variability, whereas AVs produce more structured, though still oscillatory, dynamic patterns. This underscores the necessity of carefully managing AV integration strategies to mitigate unintended amplification of traffic oscillations.In conclusion, this dissertation establishes a systematic and integrated framework for modeling and understanding mixed traffic dynamics, effectively capturing both the stochasticity inherent in human driver behaviors and the nonlinear dynamics introduced by diverse AV control strategies. The outcomes of this research provide critical insights and methodological contributions that support the development of advanced traffic modeling tools and simulation approaches. Ultimately, these findings contribute to a comprehensive understanding of traffic dynamics in mixed-vehicle contexts, informing future traffic management strategies and supporting the broader transition towards increasingly automated transportation systems.
■590 ▼aSchool code: 0262.
■650 4▼aTransportation
■650 4▼aEngineering
■650 4▼aAutomotive engineering
■650 4▼aEnvironmental engineering
■653 ▼aAutomated vehicle
■653 ▼aCar following
■653 ▼aFundamental diagram
■653 ▼aHuman driven vehicle
■653 ▼aMixed traffic
■653 ▼aTraffic hysteresis
■690 ▼a0709
■690 ▼a0775
■690 ▼a0537
■690 ▼a0540
■71020▼aThe University of Wisconsin - Madison▼bCivil & Environmental Engr.
■7730 ▼tDissertations Abstracts International▼g87-01B.
■790 ▼a0262
■791 ▼aPh.D.
■792 ▼a2025
■793 ▼aEnglish
■85640▼uhttp://www.riss.kr/pdu/ddodLink.do?id=T17358499▼nKERIS▼z이 자료의 원문은 한국교육학술정보원에서 제공합니다.


