Design and implementation of a pattern tracking system with visual control based on images for an UAV in indoor environments.

Abstract

For target tracking when UAV's operate in indoor environments, they present difficulties when using the GPS signal. One of the solutions to this type of problem is visual feedback through a camera on board the aircraft. In this work we developed two types of controllers a classic PID and a controller that uses the kinematic model of the UAV that allows to make the tracking of a defined pattern in space within indoor environments, both controllers receive a visual feedback through a camera on board the UAV that by means of a calibration estimates the distances of the UAV with respect to the pattern in each of its axes, finally a comparison of results is made and it is determined that the controller with kinematic model presents an error less than 5% for each axis in trajectories greater than one meter being the most optimal in this work.