赵靖

入学时间：2011级

答辩时间：2014年

论文题目：提升道路通行能力时空协同优化控制理论与方法

中文摘要

摘要

城市交通拥堵被认为是世界性的难题，并日趋严重。与此同时，由于城市土地资源十分有限，诸多城市，特别是大城市的道路基础设施建设已接近饱和。因此，如何通过科学的道路交通控制管理措施，挖掘现有道路交通资源的潜力，适应人们日益增长的出行需求，是交通研究人员面对的严峻挑战。本文基于时空协同优化的思想，从节点、连线、通道和网络等不同层面，对于信号控制和道路空间的交通流通行权控制理论进行了深入探讨，建立了一系列提升道路通行能力的方法。

道路空间条件与交通控制条件是相互影响的，两者互为条件，在优化过程中有着极强的相关性。但由于道路空间条件在整个道路交通系统中是一个相对稳定的部分（慢变量），而信号控制相对而言是一种更易改变的参数（快变量），因此以往主要在设计阶段优化道路的空间布置，在控制管理阶段则将其作为输入条件，主要注重对交通流通行权在时间上的调整。但从系统角度分析，若将道路空间条件也纳入优化控制体系中，能使信号控制取得更好的效果。本文将交通流在道路空间和时间上的通行权分配问题纳入到统一的研究框架内，针对提升道路通行能力这一目标，研究其协同优化控制问题。

在节点层面，传统的优化步骤一般依次确定允许及禁止流向、进出口车道、各进口车道功能、和信号配时参数。本研究基于车道控制理论，针对上述步骤，自下而上逐步提出空间结构与信号控制相结合的协同优化控制方法，分别形成车道功能动态控制、出口车道动态控制和基于中央分隔带掉头的交叉口流向控制三种时空协同优化控制方法。并分别建立了（非）线性混合整数规划模型。通过算例及仿真分析，对其交通运行效益及适用条件分别进行了评价。

在连线和通道层面，由于增加了路段要素，因此除了考虑沿线交叉口的流向管理、车道方向、车道功能和信号配时之外，重点对路段车道的优化布置进行了研究。从可逆车道的动态使用和可逆车道与其它交通管理措施的组合使用两个角度，分别针对交叉口连线和通道，形成基于动态可逆车道的连线交通控制和基于可逆车道的干道交通优化控制两项时空协同优化控制方法。并分别建立了（非）线性混合整数规划模型。通过算例及仿真分析，对其交通运行效益及适用条件分别进行了评价。

在网络层面，由于研究范围的扩展，诸如转向禁止、冲突点消除、可逆车道和单向交通等交通组织策略的组合更为多样，相互关系也更为复杂。研究从多种交通组织策略的组合优化和针对转向交通的动态控制两个角度，形成两项时空协同优化控制方法，并分别建立了双层规划模型和线性混合整数规划模型。通过算例分析，对其交通运行效益及适用条件进行了评价。

研究从中微观层面，建立了空间变量（路段各方向车道数量、交叉口转向禁止、车道功能）和时间变量（信号相位、相序、周期、绿信比、相位差）的组合优化模型，并基于算例、仿真及驾驶模拟实验评价，构筑了提升通行能力的道路交通流通行权时空协同优化控制理论与方法。

关键词：通行能力，协同优化，车道控制，信号控制，非线性规划，双层规划

英文摘要

ABSTRACT

High traffic demand coupled with unbalanced directional flows on roadways exacerbates the perennial problem of congestion, which has been one of the most depressing and challenging problems throughout the world. How to efficiently utilize the existing transportation network has been one of the most important issues faced by transportation professionals as congestion on roadways in the cities continues getting worse and the land for road construction is limited. In this paper, a series of capacity enhancement control methods for urban streets have been proposed based on the coordination optimization of lane reorganization and signal control concept, They are discussed from three different levels, including node, arterial, and network.

The roadway capacity is affected by both geometric design and signal-timing design. Since the geometric design is a more stable component in a traffic control system, its construction is harder to be varied than flexible parameters such as signal settings. Therefore, engineers have been engaged in more researches on the development and implementation of traffic signal control methods, which is called the “time management” method. However, from the system analization point of view, the geometric element should be integrated into the unified optimization framework to obtain a better control result. In this paper, a series of integrated optimization model are proposed to discuss the right-of-way assignment for traffic flow on time and space. They aim to address the following critical question that has long challenged transportation authorities during traffic management, namely: given the target transportation network and demand distribution how to select the most appropriate scheme of the reorganization of lane configurations and signal timings to enhance the facility capacity.

In the node level, the conventional optimization procedure is determining the following parameters step by step, including the permission of the movements, the layout of approach lanes and exit lanes, the lane assignment, and the signal timing. For each step, an alternative design is proposed to simultaneously optimize the geometric design and signal timing. They are dynamic lane assignment model, the exit-lanes for left-turn (EFL) control model, and the median u-turn intersection (MUTI) optimization model. These optimization models could be formulated as linear or non- linear programmings. Furthermore, their performances are evaluated through both numerical analysis and VISSIM simulation.

In the link and arterial level, the research focuses on the optimization assignment of lanes in segment. Two coordination optimization methods are proposed: operation of closely joined intersections using dynamic reversible lane control, and integrated design and operation of urban arterials with reversible lanes. The two optimization models could be formulated as a mixed integer linear programming and a mixed integer non-linear programmings, respectively. Moreover, the performances of the integrated models are evaluated through both numerical analysis and simulation.

In the network level, since the research scope is further extended, the combination of traffic management strategies, such as lane reversal, one-way street, turning restriction, and crossing elimination strategy, are more diversified and complex. Two optimization methods are proposed: network enhancement model with integrated lane reorganizationand traffic control strategies and dynamic turning restrication control. The former could be formulated as a bi-level structure with a mix integer non-linear programming problem at the upper-level, which aims to maximize the reserved capacity for the given network, and a parametric variational inequality at the lower-level, which specifies the destination distribution and route choice behaviors. And the latter could be formulated as a mixed integer linear programming. Moreover, numerical analyses are conducted to evaluate the performance the proposed model.

This paper proposes a coordination optimization method with the objective to enhance network capacity relieve congestion. The main idea is to intergrate the space elements, including number of lanes on each link, the permission of movements at intersections, lane-use assignment, etc, and the time elements, including phase plan, cycle length, green time ratio, offset, etc, in a unified framework and optimize simultaneously. The results of extensive numerical analysis and simulation show that these control methods could squeeze more capacity out of a road network and utilize the existing infrastructure more efficiently.

Key Words:Capacity, Coordination Optimization, Lane Reorganization, Signal Control, Non liner programming, Bi-level programming

王吟松

入学时间：2011级

答辩时间：2016年

论文题目：信息交互环境下城市道路交通协同控制方法研究中文

中文摘要：

摘要

为了解决城市交通问题，在宏观上，常规解决方案是增加基础设施建设以增加运能，然而由于城市空间限制，交通基础设施的建设速度赶不上车辆增加的速度，仅仅使用增加基础设施建设的方法无法完全解决日益严重的城市交通问题。在微观上，交叉口作为城市道路交通网络的节点结构，交通控制是调控交通流、提高安全性、缓解交叉口阻塞的主要措施，但传统的信息采集方式不能提供全面准确的交通控制信息，使得现状交通模式下交通控制出现响应滞后、绿灯空放等现象，影响交叉口运行效率。以车路协同系统(Connected Vehicle)为代表的信息交互技术目前已经在国内外得到了初步实现。信息交互环境一方面可以为交通控制提供更精确实时的交通需求数据，另一方面为综合考虑车辆行驶状态和交叉口信号配时进行协同优化提供了可能。本文针对城市道路交通控制亟需改善的现实需求，基于智能交通系统发展可预见的技术条件及走向，分别从连线及通道、节点两种尺度，面向常规交通、绝对优先、适时优先三类通行需求，并按照信号配时适应车辆、信号配时与车辆行驶协同优化两种思路对城市道路交通协同控制方法展开了研究。

论文首先基于现有车路通信技术背景和发展趋势，面向交通控制需求完善了车路信息交互系统架构，解决车辆定位等系统实现过程中的技术问题，搭建车载单元与路边单元，定义车辆和路侧相互发送数据的内容、格式和频率，实现车车、车路以及试验场地与实验室的数据互通，为理论研究提供技术支撑。进一步地，分析了信息交互环境下车路之间通信特征和数据特征对交通控制带来的影响，归纳总结信息交互环境下城市道路交通协同控制的基本需求、基本条件、基本控制（引导）对象以及基本优化目标，明确交通协同控制的功能体系，搭建理论研究的总体框架。

在连线及通道层面，本研究面向协调路径上双向交通流，通过信号相序、相位差和驾驶员车速调整，使得路径上尽可能多的车辆获得不停顿的连续通行权。优化车辆的行驶车速部分地取决于交叉口信号配时，其优化结果又将影响交叉口的信号控制方案，两者相互影响，互为条件。因此本研究以降低路网总延误或停车次数为优化目标，将信号配时和行驶车速两方面的参数纳入统一的优化模型，研究其求解方法，在协调路径上形成车路之间的协同优化。

在节点层面，利用信息交互环境所提供的技术条件，分别针对常规交通需求、绝对优先需求和适时优先需求提出了节点信号配时策略和方法。针对常规交通需求，利用车路信息交互环境下实时车辆数据，通过信号相位时长调整应对交通需求实时微小波动。针对以紧急救援车辆为代表的绝对优先需求，计算决策相位切换的方式和时机，保证紧急救援车在交叉口的通行效率，并减少因提供信号优先而造成对普通车流的负面影响。针对以公交车为代表的适时优先需求，利用车路信息交互环境的数据采集和通信能力，在进行交叉口信号优先配时的同时，为公交车提供车内信号灯显示、车速引导和驻站时间建议，形成公交车行驶状态与交叉口信号控制之间的协同优化。

研究最终通过实地实验对所提出的车辆行驶辅助策略、优先控制方法进行了测试验证，通过仿真实验对面向普通需求的路径协调控制及节点相位时长优化方法进行了评价分析。测试结果表明本研究所提出的方法可从总延误、停车次数、优先控制对普通车流负面影响等方面提高城市道路交通运行效率。

关键词：交通控制，车路联网，车路协同系统，协调控制，信号优先控制，实地测试

英文摘要

ABSTRACT

Traffic control is aiming at mitigating traffic congestion, improving traffic safety and energy-saving by adjusting the time and space allocation of right-of-way on the road network. Due tothe limitation of traditional traffic information content and collection methods, implementation techniques and optimization models, traffic control system does not match the increasing requirements of modern urban traffic network.With the advancement of the wireless communication technologies and the development of the vehicle to vehicle (V2V) and vehicle to infrastructure (V2V) systems, called Connected Vehicle or V2X, there is an opportunity to optimize the operation of urban traffic network by cooperation between traffic signal control and driving behaviours. This dissertation proposed a series of cooperative optimization methods for urban streets traffic control and driving assistant under the V2X concept. The basic idea is enabling the adaption between both traffic lights and vehicles. These proposed methods are designed to serve different priority demand, and are discussed from two levels, including arterial and individual intersection.

To begin with, this research designed and developed the prototype of experimental V2X system for traffic control use, improved the vehicle positioning performance by vehicle trajctory reckoning method, and defined the content and format of vehicle and traffic signal control broadcasting message set, as the technical basis for the whole research. Then, this research looked into the data and interaction features of V2X system, analyzed its impact on traffic control, and summerized the basic demand, constraints, conctrol(guide) objects and optimiaztion objectives of cooperative traffic control under V2X environment.

In the link and arterial level, this research focused on simultaneously optimization of signal phase sequence, offset and traveling speed of vehicles by speed adaption. Since the speed adaption and traffic signal timings interact with each other, this research proposed a cooperative optimization model incorporating both traffis sigal paremeters and vehicle speed, aiming at minimize total delay or total stops for the coordinated arterials. In addition, a solution method was designed in this research to reduce the calculation time.

In the individual intersection level, this research take general traffic, extreme high priority vehicle (Emergency Vehicle) and common priority vehicle(transit buses) into consideration. For general traffic, by utilization of the high resolution vehicle data provided by V2X system, astage-by-stage updated control policy was developed to respond the slight variation of traffic arrivals by cutting or extending the green time for each stage. For emergency vehicles, this research proposed a dynamic traffic signal preemption strategy to reduce emergency vehicle’s delay at intersection and meanwhile appropriatelyswitch the traffic signal phase to achive less negative impacts on general traffic. For transit signal priority, this research proposed a cooperative bus priority method based on the communication capability of V2X system. Through communications between buses and roadside infrastructure, buses will be able to proceed unimpeded through an intersection not only by changing the traffic light, but also by notifying the driver an optimal driving speed or an appropriate time to leave a bus stop.

Finally, these aforementioned methods were evaluated by numerical analysis, simulation and field test with the proposed prototype of experimental V2X system in the real traffic environment. Test results indicate that these proposed traffic control methods can improve traffic efficiency of urban streets in terms of delay, stops and the signal priority’s negative impacts on general traffic.

Key Words: Traffic control, Connected Vehicle, V2X, Traffic Signal Coordination, Traffic Sigal Priority, Field Test

郝妍熙

入学时间：2011级

答辩时间：2017年

论文题目：车路协同环境下专用路权公共交通优化控制理论研究

中文摘要

摘要

随着人口增长与城市化进程的持续发展，城市交通拥堵已成为世界性难题，并有愈演愈烈的趋势。一方面，城市土地资源日益紧张，通过拓建基础设施来缓解道路拥堵难以实现。另一方面，优化面向普通车辆的道路交通控制管理措施虽然可以提高道路利用率，却依旧不能有效应对机动车迅速增长带来的道路资源供需不平衡。公共交通因其较大的载客量与人均通行能力，被普遍视为缓解交通拥堵的有效手段。全世界许多大中城市均建有大规模的公共交通系统（常规公交、快速公交、有轨电车等），其中公交专用道、专用路权的中运量交通（快速公交、有轨电车等）因其运量大、设计车速高和不受普通车辆干扰的特点而广受大中城市青睐。但在实际应用中，一方面公交专用路权的划分占用了道路资源，尤其是在城市道路瓶颈点——交叉口范围内会较大地降低普通车辆的通行能力，可能加剧整个交通系统的拥堵程度；另一方面，交叉口交通信号控制可能对公交车造成延误，公共交通专用路权需要与信号优先控制协同运作，否则专用路权给公交运行带来的效益会被交叉口延误抵消。但传统交通环境下公交检测手段（断面检测）与交叉口管控手段（单点优先控制）的局限，具有专用路权的公共交通交叉口优先控制方案无法精准实施。信号优先方案对交叉口信号配时进行改动时，会造成不同流向车道利用率不均，导致专用路权公共交通在现实中的运行效果并不理想，控制方法有待提高。因此，本文针对专用路权公共交通及其沿线交叉口亟需提升运行效率的现实需求，以时空集成优化与提高通行能力为理念，结合近年来在产业界逐渐成熟的车路协同技术为公共交通优化控制带来的新技术条件，研究并探讨车路协同环境下专用路权公共交通优化控制理论。

本研究首先从空间方面介绍了我国专用路权公交优先的分类与发展状况，然后介绍了现有专用路权公共交通优先控制方法。随后，分析了车路协同环境下的交叉口信号控制概况，并根据车路协同现有技术，将公交车行驶特性与其进行融合，设计构建面向公共交通优先控制的车路协同实验平台环境、定义车-路信息交互内容及格式。最后阐明了专用路权公共交通控制中存在的问题以及车路协同技术从控制环境层面所带来的优势。

公共交通运行过程受多方面的因素影响，这些因素相互独立，但其在公共交通协同优化控制中的作用结果相互影响，并影响协同优化控制效率。本研究解析了专用路权公共交通车辆运行全过程，提出公共交通过程优化控制理念，定义专用路权公共交通过程优化控制参量并给出优化方法，初步建立专用路权公共交通过程优化控制理论框架。

公共交通优化控制中，公共交通运行的准点性或车头时距稳定性重要过其绝对车速。本研究针对公共交通这一特点，研究了优先需求量化方法、优先有效性判定方法等协调控制中需要使用的基本方法。并基于这些基本方法，初步建立了专用路权公共交通多交叉口协同优先控制方法。

建立公共交通专用路权条件下信号配时-车道功能协同优化方法。由将交叉口信号配时、公交优先及车道功能动态划分等交通组织与控制要素协同考虑，建立协同优化方法。充分利用交叉口信号配时与车道功能分配之间的耦合关系，建立了优化空间更广的协同优化方法。从可变车道功能及交叉口多优先请求响应两个方面，形成四种协调优化方法：1.公交专用道专用路权适时复用，将公交专用道的进口道部分作为可供普通车辆左转车辆使用的可变车道。该方法利用公交车与普通车辆行驶流向不同这一特点，提高公交专用道使用效率，即提高公交车与普通左转车辆的通行能力，从而有益于整个交叉口的运行。2.基于动态可变车道的公交优化控制，将单根或多根普通车道用作可变车道，在不同信号灯控制相位供不同流向的车辆使用。该方法也可提高单根车道的使用效率。本方法通过交叉口配时与车道功能动态划分的组合优化方法，在向公交车辆提供优先的同时，减小对于普通车辆的影响。3.单交叉口多请求优先。在公交专用道专用路权适时复用基础上，利用可变车道优势，当交叉口有多辆公交车同时发出优先请求时，将车道分配与交叉口信号配时结合起来，建立了面向多公交优先请求的交叉口信号配时优化与车道功能动态划分的协同优化方法。该方法可以响应多辆公交车优先请求，并减小对于其他车辆的影响。4.在普通交通协调控制条件下，建立综合考虑公交优先和普通车辆协调控制的优化模型，在保证公交优先的同时，对普通车辆协调控制进行优化，减小公交优先对于普通车辆协调控制的负面影响。

本研究建立基于车路协同环境这一新技术的公交优先控制环境；分析专用路权公共交通运行特征并初步建立专用路权公共交通过程优化控制理论框架；并基于此理论框架，将交叉口信号配时、公交优先及车道功能动态划分等交通组织与控制要素协同考虑，建立时空协同优化方法；最后对本研究所提出的专用路权公共交通优先控制理论与方法进行测试验证。测试验证结果表明，本研究所提出方法不仅可以提高专用路权公共交通通行效率、安全性与经济性，同时可以降低对于普通车辆通行效率的负面影响，从而提高城市道路运行效率。

关键词：交通控制，公共交通，车路协同系统，协调控制，专用路权，信号优先控制

英文摘要

ABSTRACT

Urban traffic congestion has been a difficult problem to solve for relevant authorities and scholars. Most of the congestions are caused by the imbalance between supply and demand. On one hand, the shortage of available land resources can’t afford more road or road widening. On the other hand, the traffic control of general traffic can increase capacity of urban roads. but in the long run, traffic control still not able to solve the imbalance between supply and increasing demand. Providing highly reliable public transit services has been widely realized as an effective way to reduce traffic pressure on congested urban roads. Most of the large and medium-sized cities has implement traffic priority systems. To improve the reliability of transits, transit signal priority (TSP) and dedicated bus lane (DBL) were proposed to optimize the allocation of space and time resources on road. The DBLs provide dedicated lanes to avoid mixing buses with general traffic, and with TSP, buses can pass the intersection with a requested priority signal phase. Due to their low-cost feature and effectiveness for mitigating traffic congestion, DBL and TSP have become pervasive throughout the major cities in the world. Especially in China, among the 70 major cities, more than 80% of them have DBLs and 40% of them have bus rapid transit (BRT) or trams that are often provided with TSP. However, the effectiveness of control still need enhancement. The detection and control methods are fixed，which means transit can only be detected at a certain position and controlled by intersection controllers. The precision of detection and control is not enough for precise control, which lead to failure of priority response. Therefore, it is necessary to develop advanced and effective transit optimizing theories and methods.

Firstly, this research introduces the classification and developing status of transit priority implementation in China. Analysis the signal control under connected vehicle environment, and introduce the environment to transit priority to establish a platform and environment for transit priority. Illustrate the insufficiency in the existing control system and the advantage brought by connected vehicle environment.

Proposes a series of control parameters in the process control of transit priority, the method of priority time quantization and the method of estimating the effectiveness of the signal priority control method.

To overcome the drawbacks of transit priority, this paper focused on providing bus priority on dedicated bus lanes as well as maximizing the capacity of intersections. This research focuses on combining transit priority, lane assignment, traffic signal control and traffic management in the method. Traffic signal control and lane assignment effect each other and also provide wider optimization space. This research focuses on the time-space coordinated control and proposed four coordinated optimization control method. Utilizing the concept of IBL and proposed an intermittent bus lane for left-turn (IBLFL) method to increase the capacity of signalized intersections, in which the dedicated bus lane in the middle of the road can be intermittently used for general left-turn traffic. The IBLFL method considered both operation constraints of buses and general traffic by coordination among bus operation, intersection main signal and the pre-signal at IBLFL entering area. Test results drawn from time-space diagram and simulation evaluation show that the proposed IBLFL approach performed well in terms of increasing the capacity of intersection and reducing average delay for both transit and general traffic. Another research presents an integrated optimization (IO) model to improve the performance of intersections with dedicated bus lanes. The IO model integrated geometry layout, main-signal timing, pre-signal timing and transit priority. The test results indicate that the proposed model can increase capacity and reduce delay of general traffic when providing priority to buses. Based on the IBLFL methods, this research proposes a method responding multi-priority request. The method integrated dedicated bus lane assignment, main signal control and pre-signal control. With this method, buses from conflict directions can receive priority and the adverse impact to general buses is minimized.

This paper proposes a coordinated transit optimizing control based on dedicated bus lane. the method focused on the whole operation process of transits and established a series control parameters. The mean idea of the research is to integrate time (main signal control, pre-signal control, travel speed of transit, and etc.) and space (dedicated bus lane assignment and general traffic lane assignment) factors in a unified framework and receive a global optimum, which benefit both transit and general traffic.

Key Words: Traffic control, public transport, Connected Vehicle, V2X, Traffic Signal Coordination, Traffic Sigal Priority

陈文卿

入学时间：2011级

答辩时间：2017年

论文题目：面向生态型驾驶的平面道路交叉口车路协同控制策略研究

中文摘要

摘要

近年来，环境污染和能源危机问题愈发引起公众的关注。美国官方数据显示各类交通出行行为消耗了百分之七十五的汽油，并贡献了第二大的碳排放量。研究显示不良的驾驶习惯会导致大量不必要的油耗和尾气排放。为了降低能源的浪费和对环境的污染，培养良好的驾驶习惯是当务之急，而生态型驾驶则是一种绿色驾驶行为。生态型驾驶的精髓在于温和地控制车速，使单位时间的车速变化最低，同时尽可能避免长时间的怠速，以到达节能减排的目的。

自从车路协同技术被提出以来，利用短程通讯技术获取车辆的实时车速、位置等信息以及交叉口相关参数，使得优化车辆行驶行为而实现生态型驾驶从理论上已成为研究热点。

本课题旨在提出一系列面向生态型驾驶的城市道路交叉口车路协同控制策略，以在满足不同类型交叉口的效率与安全需求的基础上，实现对环境较为友好的驾驶目标。

控制策略的应用类型分为两类：第一类是侧重于通行效率的场景，即城市信号控制交叉口。控制对象由最小单元的单车控制转向车队控制；另一类则是侧重安全，即无信号控制交叉口（让行交叉口）和高速信号控制交叉口（规避两难区）。

控制策略的应用环境也分为两类：第一类是较为初级的阶段，即有人驾驶，需要考虑人为因素的阶段；第二类是未来的无人驾驶阶段，即无需考虑人为因素的阶段。

在正式提出控制策略之前，本课题利用目前较为流行的VT-Micro微观排放模型分析了影响车辆油耗和排放的关键因素。研究发现，在不限制加/减速时间的前提下，加/减速对油耗或排放的影响远远大于速度。同时，研究分析了上述控制场景对油耗和排放的影响，发现信号阻滞或不可穿越间隙会使车辆较为激进的停车，从而产生大量的油耗和排放；此外，长时间的怠速也会带来不必要的资源浪费。

在分析了油耗和排放的影响因素后，本课题针对各种应用场景提出了控制策略。首先，本课题提出了针对信号交叉口的单车动态车速控制策略。该策略不仅考虑了目标车辆的行驶状态，同时也考虑了信号信息和下游车辆的影响。算法将加/减速替代车速作为车速控制的目标参数。针对此算法，本课题提出了若干算例，结果显示在不同的测试场景中，算法可以有效地降低目标车辆的油耗与尾气排放。研究同时对不同初始车速下的控制效果进行了灵敏度分析。测试结果表明在不同的场景下，初始车速的大小与控制结果呈不同类型的相关关系。由此可以帮助策略实施者在距离交叉口较远的位置实施初步的车速控制。

其次，本课题进一步提出了信号交叉口的车队生态型车路协同控制策略。研究同样利用加/减速度替代车速作为目标参数，以最大化地清空车队内部车辆为目的，同时尽可能降低车队停车等待的次数。算法同时考虑了信号状态和下游车队所带来的影响。当考虑一队由服从指令和不服从指令的车辆组成的车队时，算法会根据车队中两种车辆的排列组合重新划分新的车队。最后，研究所提出的算例验证了算法的有效性。

再次，针对无控（二路让行）交叉口，研究提出了基于间隙的车辆车速控制算法。该算法将可用间隙视为绿灯时间，而将不可用间隙与车辆行驶自身长度所需的时间之和视为红灯时间。算法考虑了下游车队及动态间隙的影响。算法类似利用加/减速度替代车速作为油耗目标参数。研究提出若干算法以验证模型的有效性。结果显示算法使车辆成功规避冲突点并降低油耗。

最后，针对高速信号控制交叉口的两难区问题。本课题提出的控制策略在保护高速车辆避免触发两难区效应的同时实现了生态型驾驶的目标。考虑到在高速行驶过程中，人为操作很难在极短的时间内将车速改变至一个精确值，因此本算法提出一个速度区间以替代精确的车速作为目标范围。此外，算法限制了加/减速时间以避免长时间变速。算例结果验证了算法的有效性。

综上所述，面对各类交叉口的车路协同控制算法在理论上实现了生态型驾驶的目标。由此为进一步在现实场景中实施车路协同控制算法提供理论基础。

关键词：车路协同、生态型驾驶、车速控制、交通工程

英文摘要

ABSTRACT

In recent years, public attention to environmental pollution and energy shortage is growing rapidly. In the United States, the transportation sector uses up nearly 75% of petroleum and emits the second largest carbon dioxides. Research findings indicate that bad driving behaviors constitute the major contribution to carbon emission and petroleum consumption. To reduce resource wastage and alleviate environmental problems, alternations of those inappropriate driving behaviors are suggested and referred to as eco-driving.

Leveraging DSRC technology, approaching vehicles’ speeds, positions, and other information associated with signal data could be obtained at a signalized intersection, which makes it possible to optimize vehicle movement status realize eco-driving.

This paper proposes a series of connected-vehicle control strategies to satisfy the needs of safety and efficacy with different control models of intersections. Besides, these algorithms also make traffic environmental friendly.

Firstly, this paper develops a dynamic vehicular speed control algorithm towards eco-driving at a signalized intersection using the connected-vehicle technology. The proposed algorithm considers the running status of the target vehicle as well as the impact of the downstream vehicles (if exists) and the signal control in real-world traffic environment. Acceleration/deceleration profile, instead of speed trajectories, is optimized for speed guidance. Illustrative examples are provided to validate the proposed algorithm. Results indicate that the proposed control algorithm is effective to minimize the fuel consumption and emission of the target vehicle under various test scenarios. Sensitivity analyses with respect to the impact of initial speeds of the target vehicle on the speed control performance are also conducted.

Secondly, the research advances an eco-driving speed control algorithm for a platoon of vehicles at a signalized intersection. Acceleration/deceleration profile, instead of speed trajectories, is used as the optimization objective to minimize idling and to maximize the chance of vehicles clearing the intersection during the green light. Both the running status of the target platoon and the impact of the downstream platoon are considered in the proposed algorithm. Considering a platoon composed of vehicles obeying or disobeying the speed guidance, the proposed algorithm regroups vehicles into new platoons according to their permutations. Three illustrative examples are presented to validate the proposed algorithm.

Thirdly, a gap-based automated vehicular speed control algorithm towards eco-driving at an unsignalized intersection is proposed. The algorithm takes the acceptable gaps as green times and unacceptable gaps as well as vehicle lengths (measured by time) as red times. It also considers the running status of the target vehicle as well as the impact of the downstream vehicles (if exist) and the dynamic gap-acceptance conditions in the real-world traffic environment. Acceleration and deceleration profiles, instead of speed trajectories, are optimized for speed guidance. Illustrative examples are provided to validate the proposed algorithm. Results indicate that the proposed control algorithm is effective to prevent conflicts at the unsignalized intersection while minimizing fuel consumption and emission of target vehicles under various test scenarios.

Finally, the research develops an automated speed guidance algorithm to prevent high-speed vehicles approaching a signalized intersection from being trapped into a dilemma zone while maintaining their eco-driving profiles. Considering vehicular speed fluctuation, a speed range instead of an accurate speed is used for speed guidance. In addition, a vehicle’s acceleration/deceleration time is constrained to prevent long time active speed alteration. Results from illustrative examples validate the effectiveness of the proposed algorithm.

In summary, various connected vehicle control algorithms towards corresponding control type of intersections realize eco-driving theoretically. So it is useful to support theoretical base for applying these algorithms in real world.

Key Words:connected vehicle, eco-driving, speed control, traffic engineering

王嘉文

入学时间：2011级

答辩时间：2017年

论文题目：基于交通态势的城市道路网络交通非常态主动控制基础问题研究

中文摘要

摘要

随着经济与社会的发展，城市机动车保有量与交通出行需求不断增加，城市道路网络交通拥挤现象对经济、环境均存在不利影响。由于大城市土地资源有限，社会与经济因素限制了交通基础设施的建设，如何科学有效地应用交通管理与控制措施，提高现有交通设施的使用效率，适应大城市日益成长的交通需求，成为对策现阶段城市交通拥挤的重要研究内容。城市道路网络中的随机扰动是造成网络交通供需结构性不匹配，导致大规模交通拥挤，形成非常态条件的主要原因。因此，对策随机扰动成为交通控制研究领域亟待解决的关键问题。本文基于交通态势采集、分析技术，梳理了城市道路网络主动控制机制，从对策静态扰动与动态扰动两个角度，深入探讨了非常态条件下的城市道路网络主动控制理论，建立了一系列提升城市道路网络运行效率的主动控制方法。

网络交通态势与交通信号控制两者的关系密切，二者互相影响，在交通控制理论中，交通态势是重要的输入条件之一。从系统角度分析，针对具有随机性与突发性的非常态控制问题，将通过交通检测器采集、分析、处理所得的实时交通态势信息纳入控制机制中，可以使非常态条件下的交通控制取得更好的效果。本研究提出了可行的基于动态检测器的路段行程车速估计方法与城市道路网络交通态势分析方法，为主动控制提供输入条件。

城市道路主动控制问题，其本质是动态最优化问题，即复杂系统的最优化控制问题。本研究以城市道路网络控制宏观模型为基础，针对非常态条件下城市道路网络这一具体控制对象，对城市道路网络主动控制的基本概念做出讨论，明确主动控制的目标与优化对象，并厘清了非常态条件下的主动控制流程。

针对静态扰动导致的城市道路网络拥堵现象，研究了以最大化城市道路网络运行效率的边界控制策略。基于交通态势分析得到的网络流出流量方程，动态的调整控制目标，优化流入静态扰动影响的网络区域的交通流率，使受控区域内车辆数维持在效率较高的取值区间，从而达到优化网络运行效率的目标。该方法基于受控区域交通流模型，设计了反馈控制器，提高了控制的稳定性并更易于工程实现，解决了现有控制策略对非常态条件下的网络容量的过高估计导致的控制效果下降问题。通过算例及仿真分析，对其控制效果与适用条件进行了评价。

针对城市道路网路中的动态扰动，即特殊交通流现象，本研究提出一种基于优先级别的主动式优先控制方法，在保证特殊交通流通行需求的条件下减少其对常规交通流的影响。建立了基于特殊交通流优先需求强度和优先控制影响强度的特殊交通流优先级多层模糊划分模型，继而估计分级后的特殊交通流的行程时间，最后以优先级与行程时间估计值作为主要输入条件，优化特殊交通流优先控制路径与优先控制方法。通过算例与仿真分析，对其控制效果进行了评价。

本论文初步研究了城市道路网络非常态条件下的交通态势分析，非常态条件下的主动控制机制，针对静态及动态随机扰动的主动控制等基础问题，基于交通态势分析，通过算例、仿真与实际案例的分析及评价，形成了一系列以提升非常态条件下的城市道路网络效率为目标的主动控制理论与方法。

关键词：交通态势，非常态条件，主动控制，交通控制，反馈控制，优先控制

英文摘要

ABSTRACT

The amount of motor vehicles and correspondent travel demand are continuously increasing with economic and social development. The frequent occurrence of traffic congestion in urban road network has negative impacts on economy and environment. Due to the limited land resources of large cities and restrictions to transportation infrastructure construction from socio-economic factors, to apply traffic management and control measures in a reasonable and effective way, improve the efficiency of existing transportation facilities, and accommodate the growing traffic demand in big cities have become significant research contents for counteracting urban traffic congestion. Incidents (perturbation) in urban road network are the major cause for the structural mismatches of traffic supply and demand in road network, leading to large-scale traffic congestion. Therefore, counteracting incidents (perturbation) becomes the key problem needed to be solved urgently in the field of traffic control research. This thesis (i) summaries active control mechanism of urban road network based on the acquisition and analysis techniques for the network traffic situation; (ii) discusses active control theory of urban road network with the incidents (perturbation) in-depth from the aspects of counteracting static and dynamic perturbation; and (iii) establishes a series of active control method for improving traffic operation efficiency in urban road network.

Network traffic situation is one of the important input parameters in traffic control theory. Network traffic situation and traffic signal control are closely related to each other and affected by each other. From systematic viewpoint, preferable traffic control performances can be obtained by analyzing the real-time information of network traffic flow. Traffic data collected by traffic detectors are incorporated into the control mechanism aiming at control for incidents-affected region. This thesis proposes a practical road section speed estimation method based on floating-vehicle detectors and presents analysis method for network traffic situation, which can provide inputs for proposed active control theory.

The essence of active control problem in urban road network is the dynamic optimization problem. That is, the optimal control problem for complex system. Based on a macroscopic model of urban road network control, this thesis analyzes the basic conceptions of active control for urban road network. Considering urban road network with static or dynamic perturbation caused by incidents as control object, this research discusses basic concept of the urban road network active control problem.

For static perturbation in urban road network, this thesis develops a perimeter control strategy for an urban road network to maximize its throughput when static perturbation, i.e. incidents, occur. The proposed perimeter control strategy derives its effectiveness due to two characteristics. First, the control target, i.e., the optimal inflow rate of the incident-affected network, is adjusted online based on the network traffic situation, i.e. Network Exit Function, that is updated dynamically using realtime traffic data. The incident-dependent control target precludes the overestimation of the capacity of the incident-affected network. Second, the proposed perimeter control strategy applies the proportional-integral-derivative controller, which enhances control stability and is easy to implement using the adaptively-updated control targets. Simulation-based experiments are used to investigate the effectiveness of the proposed control strategy in improving the network throughput when incident occur. The results demonstrate that the proposed strategy can enhance the average speed of traffic withstatic perturbation.

Aiming at the dynamic perturbation in urban road network, i.e. the special traffic flow, this thesis proposes adegree-of-priority basedpriority control method to reduce the impact on conventional traffic flow while guaranteeing the demand of special traffic flow. A multi-layer fuzzy partition model for special traffic flow’s degree-of-priorityis established based on the specific traffic demand intensity and priority control intensity. Travel time of special traffic flow after classification can then be estimated. Finally, the degree of priority and travel time estimates are taken as the main inputs for optimizing special traffic flow priority control path and priority control method. Numerical studies and simulations are employed to evaluate control performance of the proposed method.

This thesis studies network traffic situation analysis methods for the incident-affected region of urban road network. Based on the proposed method, this thesis develops the active control mechanism for incident-affected region. Aiming at active control problem for static and dynamic perturbation, a series of active control theories and methods have been proposed to improve the efficiency of incident-affected network. Numerical studies and simulations are employed to evaluate control performance proposed by this thesis.