Autonomous Navigation for Differential Drive Robots: Grid-Based Fastslam with AMCL in ROS2

Abstract: The research aims to establish a robust autonomous navigation system for differential drive mobile robots, leveraging URDF modeling, SLAM techniques, AMCL, Gazebo simulation, and RViz2 visualization. The central objective is to enable robots to autonomously perceive their surroundings, construct accurate maps, self-localize, and navigate. The integration of URDF defines robot attributes, SLAM produces precise maps, and AMCL ensures reliable localization. Gazebo facilitates testing, while RViz2 provides real-time visualization. The outcome is an efficient navigation system empowering robots to independently navigate intricate environments. Beyond warehousing, applications span service robotics, exploration, and environmental monitoring. The research's significance lies in addressing a fundamental robotics challenge, advancing autonomous mobility across sectors. The approach's efficacy is validated through testing, with potential contributions to robotics research and real-world applications. achieved 97% mapping accuracy with a 5% deviation in simulated environments.