Highlights

  • home_icon HOME
  • Research
  • Highlights
Research on intelligent wall-climbing aerial robot platform


How are the world’s tallest skyscrapers managed? It makes us wonder. Skyscrapers are dangerous. The wind is more severe at higher levels, increasing the risk of falling. If a robot cleans the glass on skyscrapers, building management will be much easier. The present study started with an idea to use a drone robot system, and an aerial robot has been developed that can diagnose the safety of a building as well as clean it. It is considered to be key research in robot intelligence diversification.

Improvement on existing tilt rotor method using the EDF

The present study has been conducted since 2013. The research team has made an effort to miniaturize the robot and create intelligent robots. The weight lightening and miniaturization of robot platforms were necessary to respond to various situations such as narrow indoor and outdoor structures that large robots cannot access, as well as various facilities such as the surface of an aircraft, the rotor of a wind power generator and the outer walls of a building. The movement is not controlled by a person, but by a robot that detects the distance by itself and clings to the glass surface. The tilt rotor method using the existing electric ducted fan (EDF) was improved to create a stable wall-climbing robot platform combining an ordinary propeller and the tilt method. A small airframe that can perform perching, moving and detaching mechanisms was developed. In addition, the research team designed a lightweight suspension platform structure with contact surface composite materials and shapes to maximize the maintenance of the grip force while maintaining the miniaturization of the robot. It is expected to respond to various urban environments as an intelligent robot platform for movement.

Infinite possibilities of a wall-climbing aerial robot

In relation to the research on driving and attachment and detachment to and from the walls of a drone-based robot platform, the current location and pose of the robot platform can be estimated by various sensors such as a stereo camera, a marker camera and a 2D LiDAR. In addition, the research team is making efforts to develop an algorithm that enables a stable approach to the target area and the wall and location maintaining control, and an algorithm that enables a stable approach to the wall by measuring the distance to it via sensors such as a stereo camera or ultrasound and automatic clinging to the wall using sensor information. “As moving on various walls and safe takeoff and landing to and from the wall become possible, a tilt-rotor structured robot platform and rotor angle and thrust control algorithm will be integrated to make a platform that can be used for both perching and climbing on the wall” Professor Myung said. He added that this will allow the platform to be controlled by recognition of the different slopes and shapes of the walls, and explained the possibility of the research. The results of the research are likely to be used for safety monitoring purpose using a drone and for the wall cleaning and structure monitoring markets. Professor Myung expressed his gratitude to the researchers who were devoted to the research, which had many trials and errors. He is planning to increase the commercial and technical value of this research technology so that it can be commercialized through technology transfer.

Prof. Hyun Myung
2018 KI Annual Report


KAIST 291 Daehak-ro, Yuseong-gu, Daejeon (34141)
T : +82-42-350-2381~2384
F : +82-42-350-2080
Copyright (C) 2015. KAIST Institute