...
  • 主题
高级搜索  >
历史搜索 清空历史
首页> 外文期刊>Computers，environment and urban systems >Mining massive taxi trajectories for rapid fastest path planning in dynamic multi-level landmark network
【24h】

Mining massive taxi trajectories for rapid fastest path planning in dynamic multi-level landmark network

机译：挖掘大量的滑行轨迹，以便在动态多层地标网络中进行快速的路径规划

获取原文
获取原文并翻译 | 示例
           
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

This paper presents a speed up algorithm for real-time path planning based on massive trajectory data mining, which comprises three stages: preparatory work, preprocessing stage and the online fastest path query. At the preparatory stage, this algorithm constructs multi-level landmarks and divides the original road network into multiple levels accordingly. At the preprocessing stage, this algorithm first estimates the travel time of all road segments according to the real-time traffic information, then compares all taxi trajectories to extract the experiential fastest paths, and finally makes use of the multi-level landmarks to obtain the rough fastest paths for all landmark pairs. At the online fastest path query stage, the server side first responds by returning a rough fastest path based on the preprocessing result, and then refines it by iteratively introducing the experiential fastest paths. A series of experiments are made to compare the proposed algorithm with the other three algorithms. Experiments indicate that the proposed algorithm has the ability to find more time-saving paths in response to client requests. More importantly, because this algorithm is capable of ensuring the fast completion of pre-computation on the server side, it has an evident advantage in the time cost of the online fastest path query compared with the other three algorithms, which is particularly suitable for the online optimal path query from a larger number of end users.
机译：本文提出了一种基于海量轨迹数据挖掘的实时路径规划加速算法，包括准备工作，预处理阶段和在线最快路径查询三个阶段。在准备阶段，该算法会构建多层地标，并将原始道路网络相应地分为多个层。在预处理阶段，该算法首先根据实时交通信息估算所有路段的行驶时间，然后比较所有滑行轨迹，以提取出体验最快的路径，最后利用多级路标获得路标。所有地标对的最快的粗略路径。在在线最快路径查询阶段，服务器端首先根据预处理结果返回一个粗略的最快路径作为响应，然后通过迭代引入体验最快的路径来对其进行优化。进行了一系列实验，以将所提出的算法与其他三种算法进行比较。实验表明，所提出的算法能够响应客户请求找到更多节省时间的路径。更重要的是，由于该算法能够确保在服务器端快速完成预计算，因此与其他三种算法相比，它在在线最快路径查询的时间成本上具有明显的优势，特别适用于来自大量最终用户的在线最佳路径查询。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱：kefu@zhangqiaokeyan.com

京公网安备：11010802029741号 ICP备案号：京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有

  • 客服微信

  • 服务号