城市交通诱导与控制系统一体化集成理论与技术研究-车辆工程专业论文.docxVIP

摘 要 随着我国经济的快速发展，城市交通问题越来越严重，并已逐步成为全社会关注 和讨论的重点。解决城市交通问题的传统措施已无法满足现代城市的发展要求。鉴于 城市交通的复杂性和传统交通管理方法的局限性，本文运用多智能体技术、博弈论、 蚂蚁算法、遗传算法等，对城市交通诱导与控制一体化集成的优化理论技术进行研究。 首先，分析了国内外城市交通诱导与控制系统一体化集成的相关研究，对智能体 的特征属性、结构模型以及多智能体间的协作与协调方法等进行了分析。 其次，对城市交通流诱导系统（UTFGS）和城市交通控制系统（UTCS）的相关 理论和技术分别进行综述，重点从交通诱导与控制一体化的时空特性、两系统之间的 耦合关系、集成集合的理论基础、动态交通分配四个方面阐述了 UTFGS 与 UTCS 一 体化集成的理论基础。 接着，提出了基于多智能体的动态交通诱导策略，采用“分层递阶”思想，构建 了包括中心 Agent（决策层）、区域 Agent（区域诱导层）、路段 Agent（执行层）的三 层动态交通诱导系统结构的框架。建立了路段 Agent 路径寻优模型，并根据蚂蚁寻径 原理给出了路段 Agent 路径寻优方法。 最后，通过多智能体技术将交通诱导系统和交通控制系统进行一体化集成，构建 了基于多智能体系统的交通诱导与控制一体化集成框架。并应用博弈论知识，对多智 能体系统的协调机制进行研究，建立了博弈模型，采用逆向归纳法获得了子博弈精炼 NASH 均衡，最后进行了博弈模型求解算法的分析。 实例应用表明，论文提出的交通诱导与控制一体化集成理论技术能够较好地提高 城市交通服务水平，对于城市交通管理实践具有指导意义。 关键词：交通诱导；交通控制；一体化；多智能体；博弈论；蚂蚁算法；遗传算 法 Abstract With rapid developing of Chinese economic, urban traffic problems were getting worse, and it gradually become focus of the whole society’s attention and discussion. The traditional solving measures can not meet the develop requirement of modern city. In the light of the complexity of urban traffic and the limitation of traditional traffic manage measures, with use multi-agent technique, game theory, ant algorithm and genetic- algorithm, this paper have done some research on the integration of traffic guidance and traffic control. First, analyzed the related research about the integration of traffic guidance and traffic control on and abroad. The agent’s feature, Structural model and the coordinate method between agents were also analyzed. Secondly, summarized the related theory and techniques of Urban Traffic Flow Guidance System (UTFGS) and urban traffic control system (UTCS) respectively, expound the research on base of UTFGS and UTCS’s integrate emphasis on four aspects: the time-space characteristics on integration of UTFGS and UTCS, the coupling relationship between the two systems, the theoretical basis of the integration and dynamic traffic assignment. Thirdly, advanced an dynamic traffic guidance strategy based on mult