分层控制算法在过饱和交通干线控制中的应用

Hierarchy Control Algorithm and Its Application in Urban Arterial Control Problem

提出一种启发式分层控制算法框架来解决过饱和城市交通干线的控制问题。整个控制器由控制层和协调层组成,在控制层,引入主线延误和次线延误的概念,将控制问题描述为一个冲突的多目标控制问题,提出一种新的基于IPNSGA-Ⅱ遗传算法的多目标相容控制策略来产生相邻交叉口的信号配时方案;在协调层,首先确定需要进行协调的交叉口群,然后以主线延误和次线延误为目标,采用IPNSGA-Ⅱ遗传算法进行优化,实现交叉口之间的协调。提出的相容控制策略具有鲁棒性、实时性、动态性的特点,能够实现干线交叉口的实时控制。将该分层算法应用于一个七个交叉口组成的城市交通干线的交通流仿真系统进行测试,仿真结果表明,本文方法比单点控制策略能够更有效的缓解网络中的过饱和状态,减小延误。

A hierarchy control algorithm was proposed to solve the traffic signal timing control problem for an oversaturated traffic arterial. The whole controller is composed of the control layer and the coordination layer, in the local control layer, the concept of feeding delay and non-feeding delay was introduced, the control problem was formulated as an conflicted muM-objective control problem, and an IPNSGA- Ⅱ based multi-objective compatible control algorithm was used to design signal timing plan for the adjacent junctions. Then in the coordination layer, the range of the junctions required to be coordinated was defined first, and the feeding delay and the non-feeding delay were used as the objectives, and the IPNSGA-Ⅱ algorithm was used to realize the coordination between the junctions. With its robust, online and dynamic characteristics, the proposed compatible control algorithm is able to tackle the conflicted multi-objective control problem. The presented algorithm was tested in a network simulation environment consisting of 7 intersections, and the results were compared with those of the isolated control strategy. It can be concluded that the proposed method is much more effective in relieving oversaturation and reducing delay in a network than the isolated intersection control strategy.