交通网络车流量的分布式协同优化方法研究

Distributed Cooperative Methods for Traffic Flow Optimization in Intelligent Transportation Network

针对城市交通网络主干道车流量密度非连续特性,分析了交通网络车流量分段仿射模型.从最小化车流量延迟角度,给出了优化目标函数并进行凸分析.在此基础上,将城市交通网络系统描述为非完整性约束条件的非线性动态系统,利用反步法通过控制变量代换和状态转换将该系统转化为时变可控协同标准型的一般形式,并设计城市交通网络系统的分布式协同控制律.然后在城市主干道交通网络模型中,分别对协同分布式优化方法、定时控制、感应控制三种交通控制算法进行对比仿真分析,从仿真结果可以看出,本文所提出的协同分布式优化方法具有较好的性能,有效的降低了交通网络中车辆平均等待时间.

For discontinuous of density of the trunk road traffic flow, the piecewise affine traffic flow model of the flow is analyzed in this paper. From the view of minimizing traffic flow delay, an objective function, of which convex analysis is performed, is given. On this basis, urban transportation network system is described as nonlinear dynamic system with non-integrity constraints. By way of controlling variable substitution and state transition as well as back-stepping method, the common standard form of time-varying con- trollable cooperation of the system is obtained. Then we simulate distributed cooperative method, timing control algorithm and sensing control algorithm in urban trunk road traffic network model. The results show that the distributed cooperative method has a better performance in reducing average waiting time for all cars.