Automatic Matching of Multi-scale Road Networks under the Constraints of Smaller Scale Road Meshes

In this paper,we propose a new method to achieve automatic matching of multi-scale roads under the constraints of smaller scale data.The matching process is:Firstly,meshes are extracted from two different scales road data.Secondly,several basic meshes in the larger scale road network will be merged into a composite one which is matched with one mesh in the smaller scale road network,to complete the N∶1(N>1)and 1∶1 matching.Thirdly,meshes of the two different scale road data with M∶N(M>1,N>1)matching relationships will be matched.Finally,roads will be classified into two categories under the constraints of meshes:mesh boundary roads and mesh internal roads,and then matchings between the two scales meshes will be carried out within their own categories according to the matching relationships.The results show that roads of different scales will be more precisely matched using the proposed method.

A Method of Road Data Aided Inertial Navigation by Using Learning to Rank and ICCP Algorithm

As an independent navigation method,inertial navigation system(INS)has played a huge advantage in a lot of special conditions.But its positioning error will accumulate with time,so it is difficult to work independently for a long time.The vehicle loaded with the inertial navigation system usually drives on the road,so the high precision road data based on geographic information system(GIS)can be used as a bind of auxiliary information,which could correct INS errors by the correlation matching algorithm.The existing road matching methods rely on mathematical models,mostly for global positioning system(GPS)trajectory data,and are limited to model parameters.Therefore,based on the features of inertial navigation trajectory and road,this paper proposes a road data aided vehicle inertial navigation method based on the learning to rank and iterative closest contour point(ICCP)algorithm.Firstly,according to the geometric and directional features of inertial navigation trajectory and road,the combined feature vector is constructed as the input value;Furthermore,the scoring function and RankNet neural network based on the features of vehicle trajectory data and road data are constructed,which can learn and extract the features;Then,the nearest point of each track point and its corresponding road data set to be matched is calculated.The average translation between the two data sets is calculated by using the position relationship between each group of track points to be matched and road points;Finally,the trajectory data set is iteratively translated according to the translation amount,and the matching track point set is obtained when the trajectory error converges to complete the matching.During experiments,it is compared with other algorithms including the hidden Markov model(HMM)matching method.The experimental results show that the algorithm can effectively suppress the divergence of trajectory error.The matching accuracy is close to HMM algorithm,and the computational efficiency can meet the requirements of the traditional matching algorithm.

An improved probabilistic relaxation method for matching multi-scale road networks

Matching multi-scale road networks in the same area is the first step in merging two road networks or updating one based upon the other.The quality of the merge or update depends greatly on the matching accuracy of the two road networks.We propose an improved probabilistic relaxation method,considering both local and global optimizations for matching multi-scale of road networks.The aim is to achieve local optimization,as well as to address the identification of the M:N matching pattern by means of inserting virtual nodes to achieve global optimization effects.Then,by adding two attribute-related evaluation indicators,we developed four evaluation indicators to evaluate the matching accuracy,considering both geographic and attribute information.This paper also provides instructions on how to identify the proper buffer threshold during matching procedures.Extensive experiments were conducted to compare the proposed method with the traditional approach.The results indicate that:(1)the overall matching accuracy of each evaluation indicator exceeds 90%;(2)the overall matching accuracy increases by 6–12%after an M:N matching pattern is added,and by 4–6%following the addition of topology indicators;and(3)the proper buffer threshold is about twice the average value of the closest distance from all nodes.