关键词: 车辆定位, 地图匹配, 模糊模式识别, 隶属函数

Abstract: Considering the uncertainty of the vehicle location information from the sensor，a kind of algorithm based on fuzzy pattern recognition techniques is proposed to select the matching road.This algorithm doesn’t need the running vehicles’ speed and direction parameters.While using the similarity between the vehicle location track and the road network in electronic map and combining the actual travel situation of the vehicle，the algorithm gets the membership functions of the distances between the locating point and the candidate roads，the included angles between the lines linking the neighbouring locating points selected at different times and the candidate roads，the connectivity between all the candidate roads and the historical matching roads.It chooses the matching road according to the maximal membership principle.The experiment results show that the algorithm is of reliability，can be used broadly and effectively improve the precision of the vehicle location.

Key words: vehicle location, map matching, fuzzy pattern recognition, membership function