Stabilization and location of a four rotor helicopter applying vision

Abstract

In this paper, we deal with the problem of local positioning and orientation of a rotorcraft in indoor flight using a simple vision system. We apply two different approaches to obtain a navigation system for the flying machine. The first approach is based on the perspective of n-points and the second one follows the plane-based pose technique. Our aim is to obtain a good estimate of variables that are difficult to measure using conventional GPS and inertial sensor in urban environment or indoor. We propose a method to measure translational speed as well as position and orientation in a local frame