Aerial mapping based on visual for search and rescue robot application

Abstract

This research paper aims to develop aerial mapping
based on visual using RTAB-MAP for search and rescue robot
applications. Mapping these areas is crucial for locating disaster
sites that are challenging for humans to reach. Area mapping
utilizes a visual RGB-D camera as input and RTAB-Map as a data
processing system to represent a 3D map. RVIZ ROS is used for
visualizing the map in 3D format. RGB-D images are processed
using RTAB-Map to obtain several data points, including
odometry, loop closure, timestamp, x, y, z coordinates, and roll,
pitch, yaw data from the camera. The test results yielded an
optimal area mapping, where the loop closure results in finding
corresponding image data quite effectively. As the number of loop
closures increases, the area mapping quality improves. In this
study, there were a total of 450 loop closure data, which allowed
the system to create an optimal 3D map representation. The
creation of this map is expected to assist in visualizing hazardous
areas to aid in search and rescue efforts during natural disasters.