An Extracting Method of Corner Points from Laser Sensor Readings

摘要

Simultaneous Localization and Mapping(SLAM) is a hot and difficult point in recent years. There are a large number of corner structures that are generated by two walls in the environment. Corner points can easily solve Simultaneous Localization and Mapping tasks of mobile robots. This method is divided into four parts: extraction points, line segment extraction, corner point extraction and corner point feature match. First, the map environment is scanned by the laser sensor to store the information of the environment. Second, the lEFT and the least square method are used to extract the straight line from the existing points. After that, we find corner points by finding the intersection point. At last, corner points are matched by angle and distance. Result proves that the method in the article can accurately extract the corner features. This method is robust and suitable for various environments.