Mobile Robot Navigation in Indoor Environments: Comparison of Lidar-Based 2D SLAM Algorithms

In this paper, a comparative study on lidar-based 2D Simultaneous Localization and Mapping (SLAM) algorithms in Robot Operating System (ROS) is addressed. These mapping algorithms are tested and compared in both simulations in ROS and experimentally on TurtleBot3 mobile robot. The experimental results are collected under the same conditions to evaluate the performance of the considered techniques. A generalized performance metric which is based on the Iterative Closest Point method, structure similarity index and Hausdorff distance are evaluated to conduct this comparison study. In addition, the Central Processing Unit (CPU) load and memory usage of each algorithm is considered as well. The present work provides a better insight on the performances of each provided SLAM algorithm to adopt the better solution for a given application.