Model of quadrotor and algorithm of vehicle control based on neural network

This paper presents a theoretical mechanical model of an unmanned aerial vehicle — quadrocopter. It studies such basic trajectories as take-off, landing, hovering, flying in a straight line. This work also considers convergent neural network algorithm of control. Based on the results of numerical simulation a training set and а neural network were constructed. It is shown that the constructed control algorithm works satisfactorily in the case of basic trajectories. Besides this work presents results for the case when the position sensor has an error.