基于时间递推与有师学习的交通路径动态诱导  被引量:6

Traffic Route Dynamic Guidance Based on Time Recursion and Supervised Learning

在线阅读下载全文

作  者:王洪德[1] 曹英浩[1] 崔铁军[1] BAN J 

机构地区:[1]大连交通大学土木与安全工程学院,辽宁大连116028 [2]伦斯勒理工学院琼森工程中心,纽约特洛伊ny12180

出  处:《中国公路学报》2014年第1期94-99,共6页China Journal of Highway and Transport

基  金:国家自然科学基金项目(50674052);辽宁省自然科学基金项目(201202022);大连市科技计划项目(2011E15SF118)

摘  要:为提高路况信息获取的预知性和路径引导的时效性,首先综合车辆运行过程中不同时段的路况差异和人因作用,运用时间递推原理,构建交通路径动态诱导整体结构模型,比较知识库累积经验路径与实时路况信息,为驾驶人的时间最短通行路径选择提供信息支撑;然后基于有师学习思路,对路径诱导逻辑内核结构进行解析,给出局部绕行和整体绕行2种算法实现的伪代码;最后通过对不同时刻出行和相同时刻出行的实例仿真分析,验证模型及程序的可靠性。研究结果表明:该交通路径动态诱导方法可辅助驾驶人及时对路况做出正确判断,减少因经验不足和突发事件造成的时间损失,适用于安装有GPS导航的车辆。In order to improve the predictability for traffic information acquisition and the timeliness for path guiding, firstly, combining the difference of the road conditions in different periods with the human factors, and with the theory of time recursion, the whole structure model for traffic route dynamic guidance was constructed. Based on the comparison of the accumulated experience in knowledge base and the real-time traffic information, the route with the shortest time was proposed for driver's selection with real-time and effective road information. Secondly, according to supervised learning analysis, the dynamic guidance kernel structure was analyzed and the pseudo codes of the logical algorithms for local and whole rerouting were presented. Finally, with the example simulation analysis of different time travels and the same time travels, the reliability of the model and the program were verified. The results show that the method of the traffic route dynamic guidance can assist the drivers to have a correct judgment of traffic information in time, and reduce the loss of time because of lack of experience, or traffic emergencies. The guidance technology can be applied for the vehicles with GPS.

关 键 词:交通工程 路径动态诱导 时间递推 路径选择 有师学习 预测与寻找 

分 类 号:U491.14[交通运输工程—交通运输规划与管理]

 

参考文献:

正在载入数据...

 

二级参考文献:

正在载入数据...

 

耦合文献:

正在载入数据...

 

引证文献:

正在载入数据...

 

二级引证文献:

正在载入数据...

 

同被引文献:

正在载入数据...

 

相关期刊文献:

正在载入数据...

相关的主题
相关的作者对象
相关的机构对象