基于隐马尔可夫模型和支持向量机的地图匹配算法  

Map Matching Algorithm Based on Hidden Markov Model and Support Vector Machine

作　　者：赵震 顾涛 Zhao Zhen;Gu Tao

机构地区：[1]北京科技大学物流工程系,北京100083

出　　处：《起重运输机械》2020年第20期172-179,共8页Hoisting and Conveying Machinery

摘　　要：综合采用隐马尔可夫模型(HMM)和支持向量机算法(SVM),提出一种基于隐马尔科夫模型(HMM)优化的地图匹配算法。引入机器学习方法,利用支持向量机(SVM)算法对基于隐马尔可夫模型的地图匹配算法进行了优化,使用机器学习方法SVM训练状态转移矩阵,提高了状态转移概率的准确性。提出基于多重因素权重(距离、速度和方向)计算观测概率的方法,同时考虑了路段的宽度信息,提高了地图匹配的匹配精确度。基于真实数据对算法进行验证,与原始HMM算法相比,文中提出的优化算法在提高匹配精确度方面具有较好的效果,符合估计城市路径行程时间的数据需求。Based on Hidden Markov Model(HMM)and Support Vector Machine(SVM),an improved online map matching algorithm based on Hidden Markov Model(HMM)is proposed.The machine learning method is introduced.The support vector machine(SVM)algorithm is used to optimize the map matching algorithm based on hidden Markov model.The machine learning method SVM is used to train the state transition matrix,which improves the accuracy of state transition probability.A method for calculating observation probability based on multiple factor weights(distance,velocity and direction)is proposed.At the same time,the width information of the road segment is considered,and the matching accuracy of map matching is improved.The algorithm is verified based on real data.Compared with the original HMM algorithm,the proposed optimization algorithm has a good effect in improving the matching accuracy,which is in line with the data demand of estimating the urban path travel time.

关 键 词：地图匹配 隐马尔可夫模型(HMM) 支持向量机算法 路网数据 

分 类 号：TP391.3[自动化与计算机技术—计算机应用技术]