Toward rich geometric map for SLAM: online detection of planets in 2D LIDAR
Linköping University, Department of Computer and Information Science, KPLAB - Knowledge Processing Lab
Linköping University, The Institute of Technology
Article in journal (Refereed)
Journal of Automation, Mobile Robotics & Intelligent Systems(ISSN 1897-8649)(EISSN 2080-2145)
laser scanner, SLAM, LIDAR
Rich geometric models of the environment are needed for robots to carry out their missions. However a robot operating in a large environment would require a compact representation. In this article, we present a method that relies on the idea that a plane appears as a line segment in a 2D scan, and that by tracking those lines frame after frame, it is possible to estimate the parameters of that plane. The method is divided in three steps: fitting line segments on the points of the 2D scan, tracking those line segments in consecutive scan and estimating the parameters with a graph based SLAM (Simultaneous Localisation And Mapping) algorithm.
eLLIIT - The Linköping‐Lund Initiative on IT and Mobile Communications