Traffic signal hybrid control method based on iterative learning and model predictive control

DOI编号  10.7641/CTA.2020.91025
2021,38(3):339-348

 作者 单位 E-mail 闫飞 太原理工大学 yanfei@tyut.edu.cn 李浦 太原理工大学 lipu.lp@aliyun.com 续欣莹 太原理工大学

针对基于迭代学习控制的交通信号控制方法对于路网中存在的非重复性实时干扰不能进行有效处理的问 题, 本文在基于迭代学习控制的交通信号控制方法基础上, 结合模型预测控制滚动优化和实时校正的特点, 提出了 一种基于迭代学习与模型预测控制的交通信号混合控制方法. 该方法在有效利用交通流周期性特征改善路网交通 状况的同时, 可借助模型预测控制的优点对非重复性的实时干扰进行处理, 从而进一步提高交通信号的控制效率. 通过仿真实验对该方法的有效性进行了验证. 实验结果表明, 基于迭代学习与模型预测控制的交通信号混合控制 方法能够更有效地均衡路网内的车辆密度, 进一步提高了路网的通行效率. 最后, 本文还对该方法的收敛性进行了 分析.

The traffic signal control method based on iterative learning control can not effectively deal with the nonrepetitive real-time disturbance in the road network. Based on the iterative learning traffic signal control method, a mixed traffic signal control method based on iterative learning and model predictive control is proposed through combining the rolling optimization and real-time correction characteristics of model predictive control. The proposed method can effectively improve the traffic conditions of the road network by using the periodic characteristics of traffic flow and deal with the real-time disturbance through the advantages of model predictive control. Thus, the control efficiency of the traffic signals is further improved. The effectiveness of the proposed method is verified by simulation experiments. The experimental results show that the hybrid traffic signal control method based on iterative learning and model predictive control can more effectively balance the vehicle density in the road network, and further improve the traffic efficiency of the road network. Finally, the convergence of the proposed method is also analyzed.