Low Cost 2D Laser Scanner Based Indoor Mapping and Classification System

Abstract

This paper presents a low-cost indoor mapping and classification system using cheap RPLidar 2D laser scanners. In this work, a combination of two laser scanners mounted orthogonally on a trolley or backpack has been used to generate 3D map of the surveyed indoor vicinity and to classify it. The generated map has been estimated using Simultaneous Localization and Mapping (SLAM) technique while classification has been done using Random Sampling and Consensus (RANSAC) based segmentation technique. In order to completely map the indoor environment, the proposed hardware system has been required to move manually along the surveyed vicinity and all online sensors measurements have been recorded using Robot Operating System (ROS). Later, the recorded data has been playback and desired mapping and classification techniques have been applied to generate results in offline mode. Multiple tests have been conducted using proposed system and results have been found accurate and nearer to ground truth if compared using the standard manual measuring devices available in the local market.