Trajectory tracking for fixed-wing UAV using model predictive control and adaptive backstepping

This paper presents a two layer guidance and control system design for fixed-wing unmanned aerial vehicles (UAV). On the top level, the guidance system makes the airplane follows precomputed references using a novel iterative model predictive scheme, which can handle the nonlinear optimization problem by successive linearizations (starting the algorithm using a robust L1 navigation law which ensures the availability of the solution). The outputs of this algorithm are references in airspeed, flight path angle and bank angle, which are seek by the low level controller also presented in this paper. This controller uses an adaptive backstepping approach for the longitudinal dynamics, and a conventional integral-LQR controller for the lateral-directional dynamics. Simulations show good performances of the overall architecture.