路网信息互联下基于拥堵态势感知的信号交叉口群柔性调控方法

In this proposal, the dynamic traffic state and congestion level are accurately estimated and assessed by fusing the surveillance data from roadside equipment and real time vehicle trajectory. Based on the accurate estimation of traffic state, a novel signalized intersections control system will be developed to minimize road network congestion. Empirical model and analytical model will be built to describe the traffic flow characteristics. These models will be developed considering traffic dynamics, network topology, signalized intersections and driver behavior. Additionally, a hierarchical Situation Awareness model will be designed to assess the traffic congestion distribution and development. Then, a network decomposition strategy will be proposed, where each object class involved in arterial traffic will be quantified using a Situation Awareness oriented priority level index. Finally, flexible control methodology will be applied into the proposed network control system based on thoroughly estimating traffic dynamic on spatio-temporal dimension. The proposed research is subject to be more adaptive to the complex dynamic traffic situation in advantages of accurately estimation of traffic dynamic and control the signalized intersections to satisfy the equivalence of traffic demand and capacity in maximum, which is of effective to resolve the traffic congestion problem. The proposed research is expected to improve the efficiency and management of the road network.

本课题基于融合路网内泛在分布的交通检测数据和实时车辆轨迹数据，对城市区域交通状况和拥堵态势进行全时空观测、估计与感知，并建立具有多维观测和柔性调控机制的信号交叉口群控制系统新结构。综合路段交通流状态、路网拓扑、交叉口信号状态等，建立路网交通状态在时空分布上的分析模型与经验模型。提出表征路网拥堵程度的态势元、态势基、态势权评估模型，对交通拥堵形成过程中多个相关的交叉口状态和路段状态的时空分布和变化规律进行综合分析。在此基础上，以路网整体通行效率为优化目标探索路网分解和协同控制策略，对区域内的主要干道和交叉口进行协调优化与控制。最后，综合辨识路网内动态交通流状态及扰动量，建立多维状态观测下的信号交叉口群柔性控制系统。所提出的研究思路与方法能更准确把握城市交通拥堵形成、发展与消散过程的来龙去脉，最大程度上控制路网交通供需达到动态平衡，有效解决交通拥堵，提高道路通行效率和交通管理水平。

本项目将车联网技术引入到城市路网交通拥堵态势感知与调控优化方法研究中，基于车联网环境下泛在分布的车辆轨迹数据、路侧摄像头、交叉口线圈等交通监控设备的检测数据对城市路网交通状态的时空分布特性及发展规律进行全面感知，构建了交通拥堵形成与演化机理的建模方法，进而提出了对交叉口群的信号配时、信号周期以及路段上下游交叉口信号的相位差进行自适应调控，使路网交通拥堵状况在达到临界饱和值前消散。本项目的研究成果已应用于苏州高铁新城智能网联先导区、青岛智能网联示范区、武汉智能网联汽车和智慧交通应用示范区，将为缓解城市交通拥堵、提高路网整体性能提供新的理论依据和应用基础。此外，通过该项目培养了青年教师2名、博士研究生2名、硕士研究生3名。基于本课题的研究工作，项目组成员在IEEE Transactions on Intelligent Transportation System、Transportation Research Record、中国公路学报等行业知名期刊和学术会议发表论文10余篇。