Visual servoing of a micro quadrotor landing on a ground platform

Abstract

This paper presents a visual servoing methodology for landing a quadrotor-type micro aerial vehicle (MAV) on a ground platform in an indoor environment. Since space in an indoor environment is usually limited, the sizes of the MAV and the landing platform employed here have to be small. Thus, visual servoing based on two pin-hole cameras mounted on the front and bottom of a MAV is suitable for such a scenario. A powerful computational processing unit for image or control processing and a high capacity battery for charging the MAV are equipped on the ground platform. The captured images and the flying control commands are transmitted between the ground platform and the MAV through a wireless network. Visual feature extraction of the landing platform is proposed to detect a specific image pattern on the landing platform. Flying control strategies, with the cameras switching, are developed to search for and approach the landing platform. Alignment control during the landing procedure is designed using multi-level fuzzy logic control with the target position error on the image. The overall performance of the landing algorithm is validated by several online experiments with a commercialized MAV.