Map Matching using Line Segment Features for Self-Vehicle Localization in Road Environments

Abstract

Autonomous vehicle technology offers the possibility of significant benefits to social welfare such as reducing crashes, congestion, fuel consumption, and so on. Among them, collision prediction and avoidance system are especially important topics in real road scenes. To complete the collision prediction and avoidance system, we need to estimate the exact position. Here, we introduce map matching system between the detailed digital map and the features of the line segments in the image using a monocular camera. We uniformly represent curbs and white lines on the road surface as line segment features. Furthermore, digital map and surrounding environment created by line segment are matched by performing a bird’s-eye view conversion of line segment features. Finally, we verify the usefulness of this system by matching experiment between surrounding environment created by line segment and digital map.