Adaptive Longitudinal Control of UAVs with Direct Lift Control

In this paper, a nonlinear control technique based on Direct Lift Control (DLC) is proposed to control the longitudinal dynamics of unmanned aerial vehicles. The baseline controller is designed using a nonlinear dynamic inversion technique. As the effectiveness of the baseline controller depends on the knowledge of aircraft dynamic model and aerodynamic coefficients, which is difficult to be found accurately for the whole flight regime, the baseline controller is augmented with a neuro-adaptive controller. The approach uses a single layer neural network to learn the unknown dynamics and an adaptive law is employed to ensure that the UAV behaves in the desired manner. Lyapunov theory is used to show that the approximated dynamics remains bounded. Simulations results are presented to demonstrate the effectiveness of the proposed design on a six degrees of freedom model.