Autonomous target following with monocular camera on UAS using Recursive-RANSAC tracker

This paper presents a vision-based target tracking and following system using a monocular camera on an Unmanned Aerial System (UAS). The R-RANSAC tracker tracks multiple moving objects in the camera field of view and the proposed controller is capable of following a particular target selected by a user while keeping the target in the center of the image. The main contribution of this paper is that multiple objects can be tracked without imposing restrictions such as color, shape, etc. Also, the hardware test shows that the system is able to follow a target autonomously in a real-world outdoor environment. The proposed algorithm is validated on a 3DR X-8 multirotor platform using a downward facing camera.