Natural corners-based SLAM in unknown indoor environment

The simultaneous localization and mapping (SLAM) in 2D unknown environment based on the natural corners is presented in this paper. The corners are chosen as landmarks by finding the intersection points or the end points of the line segments extracted from the raw sensor data. The mapping is constructed by the Improved Extended Kalman Filter (IEKF) based SLAM algorithm, the procedure of which is analyzed for proving the lower computation cost than standard EKF- SLAM algorithm. Finally, the experiment of line extraction, corner extraction and robot localization and mapping by using P3-DX mobile robot and HOKUYO laser sensor, which shows that the mapping with natural corners can be done very well.