Home > Published Issues > 2020 > Volume 9, No. 11, November 2020 >

Safe Landing of Drone Using AI-based Obstacle Avoidance

Sedam Lee and Yongjin Kwon
Ajou University, Dept. of Industrial Engineering, Suwon, South Korea

Abstract—As the 4th Industrial Revolution being underway, many research works on drones have been actively conducted. One of the most important part of the drone technology is now dwelling on the autonomous identification and avoidance of obstacles during the flight. In usual cases, drones are following the waypoints designated before the flight by relying on the GPS signals. However, when drones are approaching the designated landing site, there might be obstacles and unforeseen objects that may critically jeopardize the safe landing of the drones. Therefore, the safe landing of the drone is becoming a very important issue. In this respect, this study investigates the possibility of applying artificial intelligence (AI) techniques to the drone, in order to enhance the safety. By integrating image sensors, AI-enabled object recognition, and drone flight control computer altogether, the drones can be more safely landed without the fear of being overturned or critically damaged due to unexpected obstacles during the landing phase of the flight. 

Index Terms—landing platform tracking, obstacle avoidance, image segmentation, artificial intelligence, two-dimensional coordinates, flight control

Cite: Sedam Lee and Yongjin Kwon, "Safe Landing of Drone Using AI-based Obstacle Avoidance," International Journal of Mechanical Engineering and Robotics Research, Vol. 9, No. 11, pp. 1495-1501, November 2020. DOI: 10.18178/ijmerr.9.11.1495-1501

Copyright © 2020 by the authors. This is an open access article distributed under the Creative Commons Attribution License (CC BY-NC-ND 4.0), which permits use, distribution and reproduction in any medium, provided that the article is properly cited, the use is non-commercial and no modifications or adaptations are made.