Adaptive Backstepping Design for a Longitudinal UAV Model Utilizing a Fully Tuned Growing Radial Basis Function Network

Abstract

Classical Radial Basis Function (RBF) neural network controller designs typically fix the number of basis functions and tune only the weights. In this paper we present a backstepping neural network controller algorithm in which all RBF parameters, including centers, variances and weight matrices are tuned online. By using a Lyapunov approach, tuning rules for updating the RBF parameters are derived and a stability and robustness analysis is presented. Additionally, we incorporate the ability to append RBF neurons such that both tracking performance and computational cost can be optimized. The condition for adding a neuron is based on a sliding RMS error window. In addition to the theoretical results, we present the controller implemented in several simulations of an auto landing sequence for a unmanned aerial vehicle (UAV) model.

Predator-prey biogeography-based optimization for parameters identification of UCAV flight control system

“…The 6-DOF nonlinear model of UCAV is illustrated in this section, which is the prerequisite for simplifying and linearizing the mathematical model (Lombaerts, 2012;Luo et al, 2011;Zhang and Duan, 2013). Figure 1 gives some information about the coordinate system of the aircraft.…”

Adaptive sliding mode robust control based on multi‐dimensional Taylor network for trajectory tracking of quadrotor UAV

“…These factors will greatly increase the instability of the system. To compensate for the disturbance caused by the external uncertainty, radial basis function NN (RBFNN) is broadly adopted as disturbance compensator and auxiliary controller on account of its strong non‐linear mapping ability [6, 23]. Lai et al [22] focus on the application of rotorcraft to autonomously load carrying and transport.…”

“…It has shown that under PID control, a helicopter UAV will successfully reject added load offsets within a fairly large range, but the external interference was not considered. In [23], an adaptive controller based on a fully tuned growing RBFNN was designed to realise auto landing sequence for an UAV. For the existing design tools, on the one hand, non‐linearity and coupling are often approximated or even neglected for the sake of simplifying system model which may not accurately measure the input and output characteristics of the system, on the other hand, some complex controllers are designed to suppress unknown non‐linear disturbances that directly increase the load of calculation.…”