随机交通网络环境下自适应最可靠路径问题

发布时间：2018-04-10 00:22

本文选题：智能交通　切入点：随机网络　出处：《吉林大学学报(工学版)》2014年06期

【摘要】：为了研究交通网络耗时最优路径选择问题,建立了随机网络环境下自适应最可靠路径问题的数学模型。首先,建立随机网络模型反映交通网络的耗时随机特性;其次,在该网络环境下定义最可靠路径策略和最可靠状态链,并且证明最可靠状态链满足动态规划的Bellman's准则;第三,构造基于动态规划的逐次逼近算法求解该问题,并且证明提出的逐次逼近算法是多项式时间算法;最后,编写基于MATLAB计算机语言的算法程序,并针对实际交通网络Sioux Falls(SF)network展开数值试验,计算结果验证了该算法的正确性和可行性。
[Abstract]:In order to study the problem of time consuming and optimal path selection in traffic network, a mathematical model of adaptive and most reliable path problem in stochastic network environment is established.Firstly, the stochastic network model is established to reflect the time-consuming stochastic characteristics of the traffic network. Secondly, the most reliable path strategy and the most reliable state chain are defined in the network environment, and the most reliable state chain is proved to satisfy the Bellman's criterion of dynamic programming.The successive approximation algorithm based on dynamic programming is constructed to solve the problem, and it is proved that the proposed successive approximation algorithm is polynomial time algorithm. Finally, the algorithm program based on MATLAB computer language is written.Numerical experiments are carried out on the actual traffic network Sioux Falls(SF)network, and the results show that the algorithm is correct and feasible.
【作者单位】：南京林业大学汽车与交通工程学院;东南大学交通学院;美国Catholic大学土木工程系;
【基金】：国家自然科学基金重点项目(U1134206);国家自然科学基金外青学者项目(51250110075,51050110143) 交通运输部西部项目(0901005C) 江苏省自然科学基金创新学者攀登计划项目(SBK200910046) 美国国家科学基金总统奖项目(CMMI-0408390);美国国家科学基金项目(CMMI-0644552)
【分类号】：U495

【共引文献】