计算机工程 ›› 2019, Vol. 45 ›› Issue (6): 259-266.doi: 10.19678/j.issn.1000-3428.0050584

• 开发研究与工程应用 • 上一篇    下一篇

基于改进HMM模型的3D景区地图匹配算法

黄娟娟,徐圆,朱群雄   

  1. 北京化工大学 信息科学与技术学院,北京 100029
  • 收稿日期:2018-03-02 出版日期:2019-06-15 发布日期:2019-06-15
  • 作者简介:黄娟娟(1992—),女,硕士研究生,主研方向为虚拟现实、地理信息系统;徐圆(通信作者),副教授、博士;朱群雄,教授。
  • 基金项目:
    北京市自然科学基金(4142039)。

3D map matching algorithm for scenic spot based on improved hidden Markov model

HUANG Juanjuan,XU Yuan,ZHU Qunxiong   

  1. College of Information Science and Technology,Beijing University of Chemical Technology,Beijing 100029,China
  • Received:2018-03-02 Online:2019-06-15 Published:2019-06-15

摘要: 针对GPS误差和地图系统误差导致3D地图定位精度较低的问题,提出一种基于改进隐性马尔科夫模型的地图匹配算法。使用面积重叠法获取候选路段,引入距离、道路宽度和历史定位点计算观测概率,利用路网拓扑信息、游客行为特性、景点与道路的相关性计算转移概率,采用Viterbi算法得出最优匹配路段。在Unity3D平台上实现景区场景3D可视化,利用校园地图进行实验验证,结果表明,该算法匹配精度达到95.4%,在3D景区导航中具有良好的实用性和较高的准确性。

关键词: 3D地图, 景区导航, 隐性马尔科夫模型, 地图匹配, Viterbi算法

Abstract: To address the low accuracy of location in 3D map due to GPS error and digital map error,a map matching algorithm based on an improved Hidden Markov Model(HMM) is introduced.This paper proposes the area overlap method to obtain the candidate road segments.The calculation of the observation probability takes the distance,road width and historical positions into account.The transition probability is computed in terms of road network topology data,the tourist’s behavior characteristics and the correlation between scenic spots and roads.Then the algorithm uses the Viterbi algorithm to get the optimal matching road segments.The presented algorithm enables the 3D visualization of the scenic scene on the Unity3D platform,and is verified by using the campus map.Experimental results show that the matching accuracy of the presented algorithm achieves 95.4%.The presented algorithm has a good performance in 3D scenic spot navigation in both practicability and accuracy.

Key words: 3D map, scenic spot navigation, Hidden Markov Model(HMM), map matching, Viterbi algorithm

中图分类号: