An adaptive trajectory control for UAV using a real-time architecture

Multirotor helicopter have generated much interest in recent years as a powerful tool both for civil and military applications. The trajectory of the aircraft is controlled by modifying the angular velocity of the rotors. Among different control techniques the PID controller finds great application due to its simplicity. It uses proportional, integral and derivative regulator to control one or more input signal. In this paper a novel real-time system, based on a real-time Neural Network, is presented to propose an alternative control technique to the PID. In order to prove the validity of the proposed approach simulations are performed through a real experimental test-bed.