Real-time globally consistent 3D grid mapping

Abstract

For path planning of mobile robots in complex unknown three-dimensional (3D) environments, an accurate 3D volumetric representation of the environment is usually required. In this paper, we present a novel method to produce globally consistent 3D grid maps in real-time, through a grid-map update strategy and an efficient data structure. We transform point clouds into the world frame provided by a visual SLAM module, then process them by the grid mapping module to produce globally consistent 3D grid maps. Experimental results in an indoor scenario demonstrate that our approach is able to update the 3D grid map efficiently and can cope with loop closures while operating in real-time.