From Simulation to Reality: Comparative Performance Analysis of SLAM Toolbox and Cartographer in ROS 2

This paper presents a comparative analysis of SLAM Toolbox and Cartographer mapping performance in both simulated and real-world environments using ROS 2. The aim of the study is to evaluate the effectiveness, accuracy, and resource utilization of each Simultaneous Localization and Mapping (SLAM) tool under identical conditions. The experiments were conducted using the Humble Hawksbill distribution of ROS 2, with mapping tasks performed in indoor environments via Gazebo simulation and physical robot tests. Results show that SLAM Toolbox demonstrated slightly more consistent map generation in environments that included human movement and small object relocations. It achieved an Absolute Trajectory Error (ATE) of 0.13 m, compared to 0.21 m for Cartographer under identical test conditions. However, Toolbox required approximately 70% CPU usage, 293 MB RAM, and a startup time of 5.2 s, reflecting higher computational demand and configuration complexity. In contrast, Cartographer exhibited slower map generation and slightly higher RAM usage (299 MB) in simulation, while requiring higher CPU load (80%) and showing greater sensitivity to parameter tuning, which contributed to less accurate localization in noise-free simulations. This study highlights the advantages and limitations of both SLAM technologies and provides practical guidance for selecting appropriate SLAM solutions in robotic mapping and autonomous navigation tasks, particularly for systems deployed on resource-constrained platforms.