城市交通日益擁堵,引發(fā)了污染、不安全以及通勤時間延長等一系列問題,給社會經(jīng)濟(jì)帶來不利影響。由于土地使用限制和環(huán)境限制,持續(xù)增加基礎(chǔ)設(shè)施建設(shè)已不能作為解決城市交通擁堵問題的方法,而是需要通過一系列的交通控制措施,最優(yōu)利用現(xiàn)有的基礎(chǔ)設(shè)施。本文將十字路口的交通燈作為控制器件,優(yōu)化紅綠燈的轉(zhuǎn)換時間可以使得城市區(qū)域內(nèi)所有車輛的總通行時間達(dá)到最少。對于一個城市區(qū)域內(nèi)的交通,反饋控制問題的復(fù)雜性要求對其基于不同的抽象級別分層地進(jìn)行求解。本文提出,基于模型的局部反饋控制器使用詳細(xì)的交通模型,在低層以較快的時間尺度選擇每一個十字路口的交通燈的轉(zhuǎn)換時間,從而優(yōu)化這一區(qū)域內(nèi)的交通流,而邊界控制在高層使用更加聚合的交通數(shù)據(jù),以較慢的時間尺度來確定進(jìn)入這一區(qū)域的車流速度,從而避免該區(qū)域擁堵。基于模型的交通反應(yīng)式策略已經(jīng)被提出了很長時間。最初,這些策略使用的模型是一些簡單的交通流預(yù)測模型,其主要是基于感應(yīng)線圈檢測器測量的交通數(shù)據(jù),而感應(yīng)線圈檢測器通常位于十字路口車輛停止線上游40米處。從那時起,基于詳細(xì)的交通流動態(tài)預(yù)測模型,許多方法相繼被提出,并采用各種數(shù)值求解算法,包括混合整數(shù)線性規(guī)劃、遺傳算法和粒子群算法。然...

【文章頁數(shù)】：161 頁

【學(xué)位級別】：博士

【文章目錄】：
摘要
ABSTRACT
List of Symbols
List of Abbreviations
Chapter 1 Introduction
    1.1 Overview of control of urban traffic
    1.2 Motivation and contributions
    1.3 Dissertation structure
Chapter 2 Preliminary
    2.1 Traffic flow theory
        2.1.1 Time-space diagram
        2.1.2 Cumulative plots
        2.1.3 Measurement interval
        2.1.4 Density, flow and space-mean speed
        2.1.5 Conservation law and fundamental diagram
    2.2 Control methodology
        2.2.1 Elements of control design
        2.2.2 Single-agent model Predictive Control
        2.2.3 Multi-agent model predictive control
    2.3 Summary
Chapter 3 Urban Cell Transmission Model
    3.1 Motivation
    3.2 Components of network
        3.2.1 Region
        3.2.2 Intersections
        3.2.3 Sinks and sources
        3.2.4 Traffic load at intersections
        3.2.5 Links and cells
    3.3 Dynamic behaviour of traffic
        3.3.1 UCTM dynamic link model
        3.3.2 Merge and diverge constraints
        3.3.3 Representation of randomness in traffic behaviour
    3.4 Performance evaluation using UCTM
        3.4.1 Delay in UCTM
        3.4.2 Queues in UCTM
    3.5 Validation of UCTM
        3.5.1 Simulation implementation
        3.5.2 Case study
        3.5.3 Validation of aggregated behaviour
        3.5.4 Validation of state trajectories of cells in different cases
    3.6 Summary
Chapter 4 Local and Coordinated Model Predictive Controllers
    4.1 Introduction
    4.2 Local MPC
        4.2.1 Local information needed by Agenti
        4.2.2 Choice of scenarios for Agenti
        4.2.3 Objective function of local agent
        4.2.4 Optimization via simulation
    4.3 Performance evaluation of Local MPC
        4.3.1 Case study
        4.3.2 Design of controllers for comparison to local MPC
        4.3.3 Performance analysis
    4.4 Coordinated MPC
        4.4.1 Scenarios
        4.4.2 Information needed by Agenti
        4.4.3 Cost function of Agenti
        4.4.4 Optimization via simulation
    4.5 Performance evaluation of coordinated MPC
        4.5.1 Simulation settings
        4.5.2 Performance analysis
        4.5.3 Green waves generation in a simple case
    4.6 Summary
Chapter 5 Stability Enhancement and Hierarchical Feedback Control
    5.1 Motivation
    5.2 Coordinated MPC with stabilization constraint
        5.2.1 Intuition about stability issues for coordinated MPC
        5.2.2 CMPC with stabilization constraint
        5.2.3 Implementation of CMPC with stabilization constraint
    5.3 Performance evaluation of coordinated MPC with stabilization constraint
        5.3.1 Detailed performance analysis of one simulations run
        5.3.2 Statistical analysis
    5.4 Hierarchical feedback control
        5.4.1 Macroscopic fundamental diagram
        5.4.2 Description of high-level controller
        5.4.3 Hierarchical control framework
    5.5 Summary
Chapter 6 Conclusions and Future Work
    6.1 Conclusions
    6.2 Future work
References
Acknowledgement
Biography



本文編號：3853931