Aircraft predictive control

MURILLO M; LIMACHE A; FREDINI P; GIOVANINI L

Congreso; XVI Congreso de Mecanica Computacional; 2012

Universidad Nacional de Salta

Abstract. This paper presents the development of an automatic control system based on model predictive control (MPC) technique. MPC consists in finding an optimal control sequence by minimizing a cost function (or objective function) that involves a linear system dynamics and also allows the inclusion of constraints. In this paper, the technique is applied to obtain the control inputs of a Cessna 172 airplane model, so it can perform different autonomous maneuvers such as climbing to reach a certain altitude or changing its velocity to a desired one. Because the aircraft model is a six degree of freedom non linear system, in order to implement MPC, we work with the linearized system at different points of the flight path. The approach allows to obtain a quadratic objective function that ensures convexity and the existence of efficient algorithms to find the optimal control. Finally, the algorithm?s performance is demonstrated through two climbing maneouvers to preset altitudes at constant speeds.