Research on Precise Location Based on Graphic Landmark Features

Position system is an important and foundational part of intelligent mobile robots, precise position is indispensable to real-time navigation of robots. With the development of computer image processing technology, image-based position technology is improving constantly. In this paper, we proposed a precise position algorithm based on pattern signpost. Firstly, the data of image is compressed to remove the redundant information, reduce the computational complexity, and improve the operation efficiency. Then we use scale invariant feature transform (SIFT) algorithm to extract the feature of image in different scale spaces. Finally we apply the triangular positioning method to achieve the precise positioning of moving robots. The experiment result proves that the proposed method is feasible for fast and accurate autonomous navigation of moving robots.