[Purpose] Robot positioning and construction of the environmental map is an important guarantee for its own task. In the unknown environment, problems of robot positioning and map creation are interdependent. Researchers have put forward the problems of simultaneous robot positioning and map creation (SLAM).
[Method] In this paper, the landmark extraction algorithm based on RANSAC method is studied for the two-dimensional environment, and the algorithm is improved according to the application requirements of SLAM problem. Finally, the mobile robot is used to experiment
on the landmark extraction and EKF-SLAM algorithm.
[Results] The results show that the algorithm of Landmark extraction and EKF-SLAM is valid.