题名 | 基于人工交通系统的城市交通信号控制系统评价研究 |
作者 | 朱凤华![]() |
学位类别 | 工学博士 |
答辩日期 | 2008-05-30 |
授予单位 | 中国科学院研究生院 |
授予地点 | 中国科学院自动化研究所 |
导师 | 王飞跃 ; 汤淑明 |
关键词 | 人工交通系统 城市交通信号控制系统 评价 出行需求模型 基于活动生成出行需求 计算实验 规则 涌现 artificial transportation systems urban traffic signal control system evaluation travel demand model generating travel demand based on activities computational experiments rule emerge |
其他题名 | A Study on the Evaluation of Urban Traffic Signal Control System Based on Artificial Transportation Systems |
学位专业 | 控制理论与控制工程 |
中文摘要 | 城市交通信号控制系统项目投资大、与道路交通安全息息相关,必须通过有效的评价方法保证系统的性能与可靠性。现有评价方法面临的主要困难在于交通场景的规模大、复杂程度高,很难在实验室里再现,而通过现场应用开展评价的成本极高甚至无法进行。人工交通系统的出现把交通仿真方法推向一个更高的层次和更广的视野,它的基本思想是利用人工社会的理论与方法,通过抽取单个车辆或局部交通行为的基本动态规律,来理解交通系统的各种状态和发展特性,以及交通系统各部分之间相互作用所“涌现”出的复杂交通现象,从而在实验室中生成现实交通系统的替代版本。基于人工交通系统的建模与实验方法,为城市交通信号控制系统的评价提供了新的有效途径。 本论文主要研究基于人工交通系统的城市交通信号控制系统的评价,研究工作的基本思路是:构建和实际交通系统平行的人工交通系统,通过研究城市交通信号控制系统在人工交通系统中的应用情况,来评价系统的实际实施效果。通过设计不同的交通场景计算实验方案,并在必要时进行多次大量的重复,我们就可以在各种具体的交通场景下观察系统的实施效果,以获得全面、准确的评价结果。论文的主要研究工作包括: 首先,在综述城市交通信号控制系统现有评价方法的基础上,提出了基于人工交通系统的城市交通信号控制系统评价方法和评价框架,并从出行需求建模和计算实验设计两方面分析了基于人工交通系统评价方法的特点和优势。 其次,根据交通系统建模的需要,研究了人工交通系统需求模型的设计思路和方法。将出行需求模型设计过程分为三个步骤,分别是人口分类、基于个体特征构造活动计划和基于活动生成出行需求,并分别研究了每个步骤的实现方法。 再次,研究了城市交通信号控制系统评价的计算实验设计方法。以天气环境、交通事故和经济活动三种不同的计算实验为例,研究了如何设计基于人工交通系统的计算实验,生成各种可能的交通场景,并分析城市交通信号控制系统在评价场景中和人工交通系统的交互过程。 最后,开展了基于人工交通系统的城市交通信号控制系统评价实验,验证了评价方法的有效性。以北京市中关村地区为物理世界的参考模型,构建了基于人工交通系统的评价平台。在评价平台的基础上,分别在正常交通流情况、发生交通事故、新建大型商业中心和异常天气情况下开展评价实验,并通过结果分析验证了评价方法的有效性。 |
英文摘要 | Urban traffic signal control system (UTSCS) has been widely applied in the world. However, there is still no effective method to evaluate its performance and reliability. The main idea of ATS is to obtain deep insight of vehicle movement and traffic evolution by extracting basic rules of individual vehicle and local traffic behavior and analyzing the complex phenomena emerged from the interactions among individuals, and furthermore, grow an artificial system to “substitute” the real traffic system in the laboratory. ATS aim to explore feasible approaches to reproduce traffic environment in the laboratory, thus they provide us a novel idea to evaluate UTSCS. The focus of this thesis is to study how to evaluate UTSCS based on ATS. The proposed evaluation method is composed of two steps. The first step is to construct one specific ATS. Then in the second step, we evaluate one UTSCS by directly applying it in the virtual traffic environment in ATS. Using this evaluation method, we can get more comprehensive and reliable result by designing evaluation experiments in many different traffic scenes and, if necessary, repeating the experiments several times. The evaluation experiments are very hard, if not impossible, to be carried out using traditional evaluation method. Main works done by this thesis are as follows: Firstly, we proposed the method and architecture to evaluate UTSCS based on ATS. We also analyzed the features and advantages of our method in two aspects, which are generating travel demand based on activities and designing computational experiments using rules. Secondly, we studied how to establish travel demand model in ATS. The process is composed of three stages, which are classifying population, establishing one whole day’s action plan of individual and generating travel demand based on individual activities. Thirdly, we studied how to design experiments in ATS to evaluate UTSCS. We designed the rule sets of three types of experiments to simulate traffic scenes in adverse weather, traffic accident, and economic activities, respectively. We also designed the interactions between UTSCS and ATS in the evaluation experiments. Finally, we performed evaluation experiments to verify the feasibility of our method. We established one specific ATS and take it as the evaluation platform. Then we demonstrated how to carry out evaluation experiments under normal traffic condition, traffic accident condition, newly-built large commercial center condition, and adverse weather condition. |
语种 | 中文 |
其他标识符 | 200418014628062 |
内容类型 | 学位论文 |
源URL | [http://ir.ia.ac.cn/handle/173211/6094] ![]() |
专题 | 毕业生_博士学位论文 |
推荐引用方式 GB/T 7714 | 朱凤华. 基于人工交通系统的城市交通信号控制系统评价研究[D]. 中国科学院自动化研究所. 中国科学院研究生院. 2008. |
个性服务 |
查看访问统计 |
相关权益政策 |
暂无数据 |
收藏/分享 |
除非特别说明,本系统中所有内容都受版权保护,并保留所有权利。
修改评论