Abstract: With the widespread application of high-definition map data usage is increasingly popular in the autonomous driving industry. This paper proposes a logical algorithm based on different road scene features to quickly identify road intersection topology connection problems for high-precision map road intersections applications in the autonomous driving industry. By using the internal connection algorithm of road intersections, various scenarios of missing connections at road intersections can be identified, resulting in effective recognition and protection for most scenarios. Experimental results based on multiple city-map data show that the logical algorithm proposed in this article is both rapid and effectively for identifying topological connection problems within road intersections, This algorithm also preventively identifies topological connection breakpoints caused by topological connection mismatch between laboratory simulation and realistic vehicle testing. The research shows promising applications on both map quality inspection, simulation, testing and other processes in the field of autonomous driving, through reducing the number of autonomous driving downgrades and improving user experience. The experimental results provides opportunities for further improvement on autonomous driving technology by solving road intersection problems and refining vehicle perception.

Key words: high precision maps, the fusion relationship between HD and SD, road intersection, map matching, topological connection