基于ADHDP的分布式交通信号实时优化控制
|
摘要
交通信号实时优化控制的目的是最大程度上减小城市道路拥堵的现象,提高城市交通运输的效率,传统的控制方法不能够最大化地利用相位的绿灯时间。为了实现城市区域化各个路口信号的协同控制,设计了新型多智能体的分布式控制方法来解决复杂的交通信号控制的问题,以车辆的最小等待时间为目标函数建立数学模型,提出一种基于执行依赖启发式动态规划的交通信号协同优化的控制算法。通过编写Matlab程序,对系统进行仿真,实验结果表明,通过动态规划的控制算法比传统的控制算法减少了车辆的等待时间,能够达到交通信号的实时优化控制,提高了人们的交通出行能力。
The purpose of real-time traffic signal optimization control is to minimize the phenomenon of urban road congestion,improve the efficiency of urban transportation,the traditional control method can not maximize the use of phase of green light time. In order to realize the coordinated control of urban regional intersection signals,a new multiagent distributed control method was designed to solve the problem of complex traffic signal control,a mathematical model with the minimum waiting time of vehicles as the objective function was established,and a control algorithm based on the coordinated optimization of traffic signals was proposed. By programming Matlab,the system was simulated. The experimental results show that the dynamic programming control algorithm reduces the waiting time of the vehicle compared with the traditional control algorithm,can achieve the real-time optimal control of traffic signals,improve people′s traffic ability.
The purpose of real-time traffic signal optimization control is to minimize the phenomenon of urban road congestion,improve the efficiency of urban transportation,the traditional control method can not maximize the use of phase of green light time. In order to realize the coordinated control of urban regional intersection signals,a new multiagent distributed control method was designed to solve the problem of complex traffic signal control,a mathematical model with the minimum waiting time of vehicles as the objective function was established,and a control algorithm based on the coordinated optimization of traffic signals was proposed. By programming Matlab,the system was simulated. The experimental results show that the dynamic programming control algorithm reduces the waiting time of the vehicle compared with the traditional control algorithm,can achieve the real-time optimal control of traffic signals,improve people′s traffic ability.
引文
[1] Werbos P J. Advanced Forecasting Methods for Global Crisis Warning and Models of Intelligence[J]. General Systems Yearbook,1997(22):25-38.
[2]林晓峰,宋绍剑,宋春宁.基于自适应动态规划的智能优化控制[M].北京:科学出版社,2013.
[3]赵冬斌,刘德荣,易建强.基于自适应动态规划的城市交通信号优化控制方法综述[J].自动化学报,2009,35(6):666-681.
[4]于德新,高鹏,杨兆升.基于遗传神经网络的区域交通控制的效果评价[J].北京工业大学学报,2010,36(4):490-494.
[5]臧利林,贾磊,林忠琴.基于模糊逻辑的交通信号控制与仿真研究[J].公路交通科技,2008,23(4):22-26.
[6]刘红红,杨兆升.基于神经网络的实时交通信号控制与仿真研究[J].交通运输系统工程与信息,2008,8(2):1009-6744.
[7]王岚君,赵燕佳,李进源,等.信号灯控制的多阶段决策模型及其前向动态规划算法[J].控制与决策,2012,27(2):167-174.
[8]陈娟,胡庆松.城市过饱和路网的非均匀自适应相容优化控制[J].信息与控制,2012,41(5):637-643.
[9]陈智,刘小明,刘文婷,等.过饱和交叉口交通信号模糊关联控制方法[J].信息与控制,2014,43(3):374-380.
[10]首艳芳,徐建闽.信号交叉口多目标动态决策模型及其优化方法[J].公路交通科技,2012,29(11):92-97.
[11]高云峰,徐立鸿,胡华,等.交叉口定周期信号控制多目标优化方法[J].中国公路学报,2011,24(5):82-87.
[12]梁骁,殷勇辉,吴颜.单交叉口信号配时的动态优化[J].公路交通技术,2010(2):125-127.
[13]温凯歌,曲仕茄,张玉梅.城市单交叉口信号多相位自适应控制模型[J].系统仿真学报,2009,21(10):3066-3070.
[14]王小川,史峰,郁磊,等. MATLAB神经网络43个案例分析[M].北京:北京航空航天大学出版社,2013.
[2]林晓峰,宋绍剑,宋春宁.基于自适应动态规划的智能优化控制[M].北京:科学出版社,2013.
[3]赵冬斌,刘德荣,易建强.基于自适应动态规划的城市交通信号优化控制方法综述[J].自动化学报,2009,35(6):666-681.
[4]于德新,高鹏,杨兆升.基于遗传神经网络的区域交通控制的效果评价[J].北京工业大学学报,2010,36(4):490-494.
[5]臧利林,贾磊,林忠琴.基于模糊逻辑的交通信号控制与仿真研究[J].公路交通科技,2008,23(4):22-26.
[6]刘红红,杨兆升.基于神经网络的实时交通信号控制与仿真研究[J].交通运输系统工程与信息,2008,8(2):1009-6744.
[7]王岚君,赵燕佳,李进源,等.信号灯控制的多阶段决策模型及其前向动态规划算法[J].控制与决策,2012,27(2):167-174.
[8]陈娟,胡庆松.城市过饱和路网的非均匀自适应相容优化控制[J].信息与控制,2012,41(5):637-643.
[9]陈智,刘小明,刘文婷,等.过饱和交叉口交通信号模糊关联控制方法[J].信息与控制,2014,43(3):374-380.
[10]首艳芳,徐建闽.信号交叉口多目标动态决策模型及其优化方法[J].公路交通科技,2012,29(11):92-97.
[11]高云峰,徐立鸿,胡华,等.交叉口定周期信号控制多目标优化方法[J].中国公路学报,2011,24(5):82-87.
[12]梁骁,殷勇辉,吴颜.单交叉口信号配时的动态优化[J].公路交通技术,2010(2):125-127.
[13]温凯歌,曲仕茄,张玉梅.城市单交叉口信号多相位自适应控制模型[J].系统仿真学报,2009,21(10):3066-3070.
[14]王小川,史峰,郁磊,等. MATLAB神经网络43个案例分析[M].北京:北京航空航天大学出版社,2013.