Validating Autonomous Ground Vehicle Navigation Across Diverse Environments: A Comprehensive Assessment of Nav2 Planners from Simulation to Hardware Implementation

The continuous advancement of autonomous systems has significantly accelerated the evolution of industrial automation, aligning with the objectives of Industry 4.0. This thesis investigates the hardware implementation of navigation systems for Autonomous Ground Vehicles (AGVs), with a particular focus on Hardware-in-the-Loop (HIL) testing. The research employs state-of-the-art Simultaneous Localization and Mapping (SLAM) techniques and navigation algorithms within the Robot Operating System 2 (ROS2) framework, aiming to enhance AGVs’ localization, mapping, and obstacle avoidance capabilities in both indoor and outdoor environments. Multiple SLAM methods were explored, including SLAM Toolbox, Cartographer, and LIO-SAM, to identify the most effective algorithm for accurate mapping and localization. Additionally, 3D SLAM was employed to generate a three-dimensional map of the environment, particularly useful in scenarios with uneven or bumpy terrain where 2D SLAM may fail. The study evaluates multiple navigation algorithms based on the Nav2 stack controllers and planners to identify the most effective approach considering the AGV’s wheel type, driving mechanism, maneuverability, and responsiveness. Simulation environments created in Gazebo were used to test and compare different controllers and planners, assessing their performance in navigating static and dynamic obstacles. Moreover, the integration of GPS technology further enhanced localization accuracy for large-scale applications. The validated SLAM and path planning algorithms were successfully deployed on both the Scout and Bunker mobile robot models in simulation, with tests conducted in both office and outdoor environments. Additionally, autonomous navigation was implemented on the hardware of the Mini Bunker model, demonstrating its capabilities to navigate autonomously in real-world indoor and outdoor environments, marking a significant step toward practical AGV applications in industrial and agricultural settings.