计及车辆信息的隐马尔可夫地图匹配优化算法  

Hidden Markov Map Matching Optimization Algorithm Considering Vehicle Information

作　　者：滕志军[1] 皇甫泽南 王安晨 Teng Zhijun;Huangfu Zenan;Wang Anchen(Northeast Electric Power University Key Laboratory of Modern Power System Simulation and Control&Renewable Energy Technology,Ministry of Education,Jilin 132012;School of Electrical Engineering,Northeast Electric Power University,Jilin 132012)

机构地区：[1]东北电力大学,现代电力系统仿真控制与绿色电能新技术教育部重点实验室,吉林132012 [2]东北电力大学,电气工程学院,吉林132012

出　　处：《汽车技术》2024年第12期15-22,共8页Automobile Technology

基　　金：国家自然科学基金青年科学基金项目(61501107);吉林省教育厅“十三五”科学研究规划项目(JJKH20180439KJ)。

摘　　要：为解决车辆在路况重叠的高架区域进行地图匹配时信号传输受到遮挡,导航易出现误匹配、输出时延增加和车道偏移等现象的问题,提出计及车辆信息的隐马尔可夫地图匹配优化算法。首先,剔除采样数据中冗余和漂移的定位点;然后,确定候选道路时生成网格索引,利用道路拓扑删除不相连道路,减少计算量、降低输出时延;最后,利用道路和车辆信息生成可信度函数,融合速度相似性改进转移概率,确定匹配路段。试验结果表明,车辆行驶至高架区域时,所提出算法匹配时间更短,时长未随路段的增多而增加,且具有较高准确率,满足车辆在三维区域的匹配需求。The signal transmission is obscured when vehicles are matched in elevated areas,and navigation is prone to mismatching,increased output latency and lane drift,etc..To address such navigation defects,this paper proposes the hidden Markov map matching optimization algorithm with vehicle information.The algorithm eliminates redundant and drifting localization points in the sampled data;generates a grid index when determining the candidate roads,and uses the road topology to delete unconnected roads to reduce the computation and output delay;generates a confidence function using the road and vehicle information,and improves the transfer probability by fusing the speed similarity to determine the matching road sections.The experimental results show that the matching time is shorter when the vehicle drives to the elevated area,and the duration does not increase with the increase of road sections;and it has a high accuracy rate to meet the matching demand of vehicles in the 3D area.

关 键 词：城市路网 隐马尔可夫模型 地图匹配 车辆信息 

分 类 号：P208[天文地球—地图制图学与地理信息工程]