Seguimiento autónomo de personas con un robot aéreo no tripulado

This work proposes to combine two main areas of the Artificial Intelligence which are: Computer vision and Motion Planning, to be incorporated in the area of aerial robotics through unmanned aerial robots. Our main objective is the tracking of people using the detection and tracking algorithms: HOG (Histogram of Oriented Gradients) and Particle Filter besides integrate it with motion planning for the autonomous flight of the quadricopter through the algorithm Rapidly-exploring Random Tree (RRT) trees and variants of it. We present the results of the algorithms as well as their effectiveness and finally the performance of the AR.Drone 2 with the proposed techniques.