Robust Forwarding Control of Aircraft Wing Rock Using Neural Networks

In this paper, we study a robust forwarding control method of an aircraft wing rock subject to uncertain dynamic systems. We adopt neural networks to estimate uncertainties and derivatives of mappings in the recursive computation with an arbitrarily small approximation error. In this sense, benefits include: 1) nonlinear dynamics is not required to be linearly parametrizable; 2) mapping filters are not necessary to be introduced into the forwarding controller to calculate analytic derivatives of mappings; 3) the so-called problem of “explosion of complexity” is efficiently tackled. In theory, the boundedness of all signals in the closed-loop system is guaranteed. Controller performances are demonstrated via simulations.