Robust autonomous flight for quadrotor UAV based on adaptive nonsingular fast terminal sliding mode control

Abstract

This paper presents a robust adaptive nonlinear controller for a flying machine represented by a quadrotor UAV (Unmanned Aerial Vehicle) in the presence of parasitic dynamics. A hybrid control strategy is used to steer the quadrotor behavior and an online adaptive methodology is introduced to improve the robustness of the adopted control technique against unmodeled dynamics. On the basis of the Newtonian formalism, the system model is formulated and decoupled into two subsystems, a position and attitude subsystems. A new Adaptive Nonsingular Fast Terminal Sliding Mode Control (ANFTSMC) is adopted to steer the quadrotor flight attitude, and a robust Backstepping Sliding Mode Control (BSMC) is used to manage the quadrotor position, contributing fast-accurate tracking despite the impact of external disturbances. The design procedure for both controllers is described thoroughly and the closed-loop stability is proved by means of the lyapunov theory. Multiple flight tests were performed and comparison is done with the backstepping controller (BS), the integral sliding mode control (ISMC), the second order sliding mode control (2-SMC) and the global fast terminal sliding mode control (GFTSMC). Simulation results show marked improvements of the quadrotor behavior with enhanced convergence time, chattering-free control efforts and strong robustness to the parasitic dynamics.