This paper presents the design, development, and implementation of a compact, cost-effective autonomous mobile robot built around a Raspberry Pi 4 microcomputer and an RPLiDAR sensor for LiDAR-based navigation. The entire software architecture is built on ROS 2 Humble Hawksbill, integrating SLAM Toolbox for 2D occupancy grid mapping and the Nav2 stack for autonomous navigation with dynamic obstacle avoidance. The system incorporates a four-wheel differential drive mechanism with optical wheel encoders and closed-loop PID control for precise odometry tracking. Key Highlights: Total hardware cost: ~219 using off-the-shelf components Raspberry Pi 4B as the central processing unit running ROS 2 RPLiDAR A1 for 360° laser-based distance measurement SLAM Toolbox for real-time map generation Nav2 stack with DWA local planner for autonomous navigation Closed-loop PID control with optical wheel encoders Average CPU utilization: 60–70% with no thermal throttling Experimental results demonstrate that affordable hardware, when paired with state-of-the-art open-source software, can yield a robotic platform capable of consistent autonomous navigation in indoor environments.
Kankati et al. (Mon,) studied this question.