Modelagem, controle otimizado por algoritmo genético e aplicação de cell-mapping em um quadricóptero

Abstract

Vehicles like the quadricopter become increasingly studied and applied in several areas, besides military applications, the drones are used for other purposes like topography, photography, cinematography and security. This end-of-course monography proposes to perform a mathematical modeling of a quadricopter, applying PID (proportional–integral–derivative) control technique to stabilize the angles and the vertical position, which were optimized by using genetic algorithms, and, finally, obtained a cell-mapping (CM) for the already controlled height loop. The mathematical modeling was performed using the Newton-Euler method, separating the plant into two interconnected subsystems where each one is responsible for describing the angular (rotational system) and linear (translational system) behaviors of the model. Two cost functions were proposed in the development of the GA so that they could be compared in relation to the gains obtained at the end of the simulation. Finally, a CM is applied to the analysis of the response of the controlled system in a map of finite states, an preliminary application of control based on the response obtained by the mapping is proposed too.