首页> 外国专利> Method for predicting future path of moving objects in road network

Method for predicting future path of moving objects in road network

机译:道路网络中移动物体未来路径的预测方法

摘要

A method for predicting a future path of a moving object in a road network is provided to search candidate traces having a partial trace similar to a query trace among past traces stored in a moving object database and increase correctness of the predicted future path by grouping the moving paths having a small deference and regarding the grouped moving paths as a similar path. A path prediction server receives future path query information including a user ID(IDentification), a starting place, a destination place, and current location information, which are trace data of a moving object on a road network, from a user(S31). The server searches past traces, which correspond to the user ID, from the starting place to the destination place via a current location in a trace database(S32). Similarity between the searched past traces corresponding to the user ID and the trace of the query information is determined(S33). Partial traces from the current location to the destination place are grouped among the traces determined as the similar trace(S34). A future path is predicted by calculating a frequency of the moving paths included in a group(S35) based on DSN(Dissimilarity based on Segment Number) and DSL(Dissimilarity based on Segment Length) functions.
机译:提供了一种用于预测道路网络中的移动物体的未来路径的方法,以在存储在移动物体数据库中的过去轨迹中搜索具有类似于查询轨迹的局部轨迹的候选轨迹,并通过对轨迹进行分组来提高预测的未来路径的正确性。偏差较小的移动路径,并将分组的移动路径视为相似的路径。路径预测服务器从用户接收包括作为道路网络上的移动物体的跟踪数据的用户ID(IDentification),起点,目的地和当前位置信息的未来路径查询信息(S31)。服务器通过跟踪数据库中的当前位置,从起点到目的地搜索与用户ID相对应的过去的跟踪(S32)。确定与用户ID相对应的搜索到的过去轨迹与查询信息的轨迹之间的相似性(S33)。从当前位置到目的地的部分迹线被分组为被确定为相似迹线的迹线(S34)。通过基于DSN(基于片段号的不相似度)和DSL(基于片段长度的不相似度)功能来计算组(S35)中包括的移动路径的频率来预测未来路径。

著录项

  • 公开/公告号KR100859566B1

    专利类型

  • 公开/公告日2008-09-23

    原文格式PDF

  • 申请/专利权人

    申请/专利号KR20070021196

  • 发明设计人 김상욱;김종대;원정임;이정훈;

    申请日2007-03-02

  • 分类号G06F19;

  • 国家 KR

  • 入库时间 2022-08-21 19:51:38

相似文献

  • 专利
  • 外文文献
  • 中文文献
获取专利

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