基于稀疏采样数据的复杂路网地图匹配算法  

Complex road network map matching algorithm based on sparse sampling data

作　　者：黄银峰 李艳红[1] 姚静怡 罗昌银 HUANG Yinfeng;LI Yanhong;YAO Jingyi;LUO Changyin(College of Computer Science,South-Central Minzu University,Wuhan 430074,China;School of Computer Science,Central China Normal University,Wuhan 430079,China)

机构地区：[1]中南民族大学计算机科学学院,武汉430074 [2]华中师范大学计算机学院,武汉430079

出　　处：《中南民族大学学报（自然科学版）》2024年第4期522-531,共10页Journal of South-Central University for Nationalities：Natural Science Edition

基　　金：湖北省自然科学基金资助项目(2017CFB135);中央高校基本科研业务费专项资金资助项目(CZY23019)。

摘　　要：地图匹配是指将GPS定位坐标正确匹配到数字地图的道路上.离线地图匹配是从记录和存储的轨迹数据中寻找车辆行驶的真实路径.采样频率和复杂路网是影响地图匹配正确率的两个最重要的因素.为了提高现有的隐马尔可夫模型地图匹配算法在复杂路网上的正确率,提出了分段验证匹配方法(SV算法).考虑到每一段子轨迹会有k条候选路径,引入一个适应度来评判候选路径与轨迹的吻合程度,选取具有最高适应度的候选路径作为局部最佳匹配路径.此外,所提算法还考虑了路段方向和车辆行驶方向的角度差和路段限速,通过这些约束条件过滤候选路段和候选点,以提高算法效率.Map matching refers to the correct matching of GPS positioning coordinates onto the roads of a digital map.Offline map matching is to find the real path of vehicle traveling from recorded and stored trajectory data.Sampling frequency and complex road network are the two most important factors affecting the correctness of map matching.In order to improve the correctness of existing hidden Markov model map matching algorithms on complex road networks,this paper proposes a segment verification matching method(SV algorithm).Considering that there will be k candidate paths for each sub-trajectory,a fitness degree is introduced to judge the degree of match between the candidate path and the trajectory,and the candidate path with the highest fitness degree is selected as the local best matching path.Furthermore,the proposed algorithm also takes into account the angle difference between the direction of the road segment and the direction of vehicle travel,as well as the speed limitations of the road segment,and filters candidate road segments and points through these constraints to improve efficiency of algorithm.

关 键 词：轨迹分段 隐马尔可夫模型 采样频率 复杂路网 

分 类 号：TP399[自动化与计算机技术—计算机应用技术]