摘要
针对交通信号灯实时控制问题,设计了一种以最小化车辆等待时间为目标的数学模型,并给出了一种能求解该模型最优解的启发式搜索第法。仿真结果显示启发式搜索算法存在求解时间长,求解效率不稳定等问题。因此在原算法基础上加入了多阶段决策优化方法,并且在各个阶段中采用了限时搜索,使得算法能在固定时间内得到结果,保证了算法的稳定性和实时性。通过实际数据仿真显示,优化后的算法对比固定周期算法减少了车辆的等待时间;对比原始的启发式搜索算法,提高了求解效率,满足了实时控制的要求。
For real-time traffic signal control issues,a mathematical model was proposed to minimize the waiting time,meanwhile,a heuristic search algorithm was given to solve the optimal solution.Simulation results show that the heuristic search algorithm suffers from being computationally complex and unstable,therefore,a multi-stage decision optimization algorithm is added,and the searches adopt in all stages subject to a time limit,which ensures a stable and real-time algorithm,and also a solution in fixed time.Simulation results based on actual traffic data show that the waiting time can be reduced in comparison with that of the fixed-time periodic control policy,the computation time can be saved in comparison with that of the original heuristic search algorithm,and it can be applied to real-time traffic control.
出处
《系统仿真学报》
CAS
CSCD
北大核心
2015年第6期1348-1356,共9页
Journal of System Simulation
基金
科技部国际合作专项(2012DFG11580)
关键词
实时交通信号灯控制
启发式搜索
多阶段决策优化
限时搜索
real-time traffic signal control
heuristic search
multi-stage decision optimization
limited search
