Autonomous Mobile Robot Mapping and Exploration of an Indoor Environment

Abstract

The main challenge for the robot is to interact with its environment. Generally, a robot can interact with its environment by achieving the necessary contact configuration and the subsequent motion required by the task. SLAM solves the challenge of robots exploring an unknown area. The robot’s goal is to collect information while exploring the environment to develop a map it, so it wants to use its map to determine its location. The current study examines multiple 2D SLAM laser-based algorithms available in the Robotic Operating System to determine the most efficient technique for creating a map in a specific current world using an Autonomous Mobile Robot (AMR) equipped with two lidar 2D scanners in a customs warehouse environment. A good map is essential for the robot to efficiently interact with its surroundings. From this viewpoint, selecting the optimal SLAM algorithm is critical since it relies on the outcome of the final map. This paper discusses, the comparison of SLAM Toolbox, G-mapping, Hector SLAM and Karto-SLAM methodologies through real-world testing in a dedicated environment, a discussion the output maps of SLAM algorithms, their advantages and disadvantages are presented. After examining and scrutinizing the maps, the G-mapping file gives a more precise map with more defined boundaries and obstructions than the rest. The G-mapping map file shows the map without slip, tilt, or skew. However, depending on the math work, the calculation approach shows that Hector Slam provides a value error lower than the G-mapping designation.