Indoor Navigation System Using the Fetch Robot

Abstract

In this paper, we present a navigation system, including offline mapping and on-line localization, for the Fetch robot in an indoor environment using Cartographer. This framework aims to build a practical, robust, and accurate Robot Operating System (ROS) package for the Fetch robot. Firstly, using Cartographer and the fusion of data from a laser scan and RGB-D camera, a two-dimensional (2D) off-line map is built. Then, the Adaptive Monte Carlo Localization (AMCL) ROS package is used to perform on-line localization. We use a simulation to validate this method of mapping and localization, then demonstrate our method live on the Fetch robot. A video about the simulation and experiment is shown in https://youtu.be/oOvxTOowe34.

On-The-Go Robot-to-Human Handovers with a Mobile Manipulator

“…A 2D occupancy map of the experimental environment is pre-computed by using a SLAM algorithm which fuses base laser scanner and depth camera measurements [17]. This map is then used to perform on-line localization using a particle filter algorithm [17]. The position of the seated human is assumed to be known within this map.…”