Visual servoing for autonomous landing of a multi-rotor UAS on a moving platform
Autor: | Iryna Borshchova, Siu O’Young |
---|---|
Rok vydání: | 2017 |
Předmět: |
Scheme (programming language)
0209 industrial biotechnology Engineering Control and Optimization Rotor (electric) business.industry Aerospace Engineering 02 engineering and technology Visual servoing Computer Science Applications law.invention 020901 industrial engineering & automation Control and Systems Engineering law Automotive Engineering 0202 electrical engineering electronic engineering information engineering 020201 artificial intelligence & image processing Computer vision Artificial intelligence Electrical and Electronic Engineering business computer computer.programming_language |
Zdroj: | Journal of Unmanned Vehicle Systems. 5:13-26 |
ISSN: | 2291-3467 |
Popis: | In this paper, a method to control a small multi-rotor unmanned aerial system (UAS) while landing on a moving platform using image-based visual servoing is described. The landing scheme is based on positioning visual markers on a landing platform in the form of a detectable pattern. When the onboard camera detects the object pattern, the flight control algorithm will send visual-based servo-commands to align the multi-rotor with the targets. The main contribution is that the proposed method is less computationally expensive as it uses color-based object detection applied to a geometric pattern instead of feature tracking algorithms. This method has the advantage that it does not demand calculating the distance to the objects (depth). The proposed method was tested in simulation using a quadcopter model in V-REP (virtual robotics experimental platform) working in parallel with robot operating system (ROS). Finally, this method was validated in a series of real-time experiments with a quadcopter. |
Databáze: | OpenAIRE |
Externí odkaz: |