Trajectory and guidance mode for autonomously landing an UAV on a naval platform using a vision approach

It is proposed a navigation system based on the unmanned aerial vehicle (UAV) onboard camera that can land on a moving naval platform. It is considered that, besides the usual sensors, the aircraft is equipped with a RGB digital camera capable of detecting markers located on the ship's deck. The proposed method uses a framework based on a Kalman filter (KF) and an efficient perspective-n-point (EPnP) approach to estimate the position, attitude and velocity of the target (landing system). This information is combined and used to generate a landing trajectory. The algorithm is suitable for real time implementation in standard workstations because of the low computational time required.