Sistemas neuro-fuzzy aplicados no controle de uma aeronave de asa fixa

Abstract

The loss of the aircraft’s control during the flight is the main cause of deaths in plane crashes. For this reason, it is of great importance to develop an efficient controller, that many times cannot be obtained through the basic techniques. In this project it is performed the control of a fixed-wing airplane using neuro-fuzzy systems, with parameters obtained through Differential Evolution and from Simulated Annealing.It is used MatLab software to obtain the controller and to evaluate the system’s performance and stability.The obtained results indicate that the controller is capable of compensating eventual system disturbances. Furthermore, the Simulated Annealing shows a shorter time of execution to find the optimum controller, by contrast, this method shows a higher standard deviation of the results, which suggests a higher randomness and imprecision than the Differential Evolution method.