MAP Matching Method Based On Geometric Feature Constraints

Map matching is used to solve the problem that the positioning result is inconsistent with the electronic map in vehicle navigation.As a kind of pattern recognition,map matching algorithm takes the road network as a template and the track point of the vehicle as the sample to be matched.The matching template with the highest similarity with the sample to be matched is the final matching result.In terms of the samples to be matched,the current map matching methods are mostly aimed at GPS(Global Positioning System)positioning data,but GPS signals are easily blocked and lost in the urban environment,and the multi-path effect is significant,and only the location information can be provided.In terms of matching template,the intersections in urban roads are complicated due to the topological relationship,which can easily lead to mismatching,thus affecting subsequent matching.In recent years,the emergence of three-dimensional sections such as viaduct,overpass and tunnel also brings impact to the traditional two-dimensional map matching.According to the actual requirements in map matching,this paper improves the traditional map matching algorithm from the aspects of data acquisition,data processing and algorithm implementation,designs a map matching method for IMU(Inertial Measurement Unit)/odometer combined navigation and positioning data,and makes full use of geometric feature constraints of road sections and roads.Paper's main work is as follows:(1)Putting forward an improved time delay map matching method of point-point,combined with the advantages of point-point matching,point-line matching,line-line matching,makes full use of roads and road geometry constraints and adopts map matching method combining rough matching and fine matching,which can revise the horizontal error and vertical error of inertial navigation system and implement the intersection of different complexity of map matching.(2)The paper adds elevation factor into the traditional map matching factor besides the basis factor of angle and distance,select sections of different heights by elevation,it solves the problem that map matching is affected by three-dimensional road section in urban road.Besides,the algorithm joins the traffic rules limit with the topological relationship,ruled out the actual impassable roads,improve the map matching accuracy,can effectively avoid driving offences because of mismatch.(3)In this project,inertial navigation and odometer are used as data acquisition sources,because inertial navigation system can provide not only accurate position information,but also course Angle,speed and other information.Compared with GPS data,positioning information is more abundant,and the optimization space of map matching algorithm is larger.the role of the odometer is to offer speed constraint of inertial navigation,reduce the divergence of inertial navigation.