3D simultaneous positioning and mapping in dark, closed spaces with an autonomous flying robot


Creative Commons License

ÇAKIR A., AKPANCAR S.

Acta Polytechnica Hungarica, vol.17, no.7, pp.7-23, 2020 (SCI-Expanded, Scopus) identifier

  • Publication Type: Article / Article
  • Volume: 17 Issue: 7
  • Publication Date: 2020
  • Doi Number: 10.12700/aph.17.7.2020.7.1
  • Journal Name: Acta Polytechnica Hungarica
  • Journal Indexes: Science Citation Index Expanded (SCI-EXPANDED), Scopus
  • Page Numbers: pp.7-23
  • Keywords: 3D, Closed spaces, Dark spaces, Flying robot, Mapping, Simultaneous
  • Open Archive Collection: AVESIS Open Access Collection
  • Isparta University of Applied Sciences Affiliated: Yes

Abstract

This study describes the steps needed to produce the required hardware and software for the 3D mapping of dark, rugged, and closed spaces such as caves, underground cities and mining pits through a DJI Matrice 100 Flying Robot Platform that has a high bearing capacity of 3600 grams and has no limit of movement in bumpy, hollowed or sloping spaces. In order to obtain the obstacle information around, during the autonomous movement of the air robot used within the scope of this study, 5 ultrasonic sensors – right, left, front, top, bottom – were used. A servomotor driven electromechanical equipment that will be used on the z-axis movement of Hokuyo UST-20LX laser sensor, which provides data in 2D, was developed to help the air robot map its environment in 3D during the autonomous movement. The control of the hardware developed and used in this study is carried out by Robot Operating System (ROS) nodes written in C++ programming language. The mapping studies were carried out by operating the robot autonomously in caves within Atabey District of Isparta Province, Turkey, at the coordinates of 37°53'41.8"N 30°32'58.5"E and 37°53'39.0"N 30°32'42.3"E. It is shown that the 3D maps produced by the system, are realistic and substantial.